Matt Turner [Fri, 14 Dec 2012 22:22:28 +0000 (14:22 -0800)]
Add ES 3 handling to get.c and get_hash_generator.py
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Matt Turner [Wed, 28 Nov 2012 20:03:26 +0000 (12:03 -0800)]
glapi: Move ARB_base_instance to the correct location
It's #107, it shouldn't be added after the #116 comment.
Reviewed-by: Fredrik Höglund <fredrik@kde.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Matt Turner [Thu, 29 Nov 2012 20:43:39 +0000 (12:43 -0800)]
mesa/tests: Add ARB_ES3_compatibility enums
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Matt Turner [Wed, 28 Nov 2012 20:12:09 +0000 (12:12 -0800)]
glapi: Add enums for ARB_ES3_compatibility
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Quentin Glidic [Wed, 28 Nov 2012 15:33:47 +0000 (16:33 +0100)]
mesa/program: Fix both Classic and Gallium build
Follow-up for
907844107252260c646aca361191ef7f121f3d23 and
3a5ad21cd3f026579eeacc25b39513711556c7ee
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57044
Tested-by: Fabio Pedretti <fabio.ped@libero.it>
Tested-by: Brad King <brad.king@kitware.com>
Andreas Boll [Tue, 27 Nov 2012 09:25:54 +0000 (10:25 +0100)]
configure.ac: fix typo in error message
Marek Olšák [Thu, 10 Jan 2013 14:23:56 +0000 (15:23 +0100)]
r300g: don't set sample positions to the pixel center if MSAA is disabled
but an MSAA resource is bound. This effectively makes the MSAA disable switch
not affect rasterization, but it still affects the alpha-to-one and
alpha-to-coverage states. This hardware just lacks a proper MSAA disable
switch.
This fixes graphics corruption in sauerbraten.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=59194
Paul Berry [Tue, 8 Jan 2013 21:30:46 +0000 (13:30 -0800)]
intel: Clean up confusion between logical and physical surface dimensions.
In most cases, the width, height, and depth of the physical surface
used by the driver to implement a texture or renderbuffer is equal to
the logical width, height, and depth exposed to the client through
functions such as glTexImage3D(). However, there are two exceptions:
cube maps (which have a physical depth of 6 but a logical depth of 1)
and multisampled renderbuffers (which have larger physical dimensions
than logical dimensions to allow multiple samples per pixel).
Previous to this patch, we accounted for the difference between
physical and logical surface dimensions at inconsistent places in the
call graph (multisampling was accounted for in
intel_miptree_create_for_renderbuffer(), and cubemaps were accounted
for in intel_miptree_create_internal()). As a result, it wasn't
always clear, when calling a miptree creation function, whether
physical or logical dimensions were needed. Also, we weren't
consistent about storing logical dimensions in the intel_mipmap_tree
structure (we only did so in the
intel_miptree_create_for_renderbuffer() code path, and we did not
store depth).
This patch refactors things so that intel_miptree_create_internal() is
responsible for converting logical to physical dimensions and for
storing both the physical and logical dimensions in the
intel_mipmap_tree structure. As a result, all miptree creation
functions interpret their arguments as logical dimensions, and both
physical and logical dimensions are always available to functions that
work with intel_mipmap_trees.
In addition, it renames the fields in intel_mipmap_tree used to store
the dimensions, so that it is clear from the name whether physical or
logical dimensions are being referred to.
This should fix the following bugs:
- When creating a separate stencil surface for a depthstencil cubemap,
we would erroneously try to convert the depth from 1 to 6 twice,
resulting in an assertion failure.
- When creating an MCS buffer for compressed multisampling, we used
physical dimensions instead of logical dimensions, resulting in
wasted memory.
In addition, this should considerably simplify the implementation of
ARB_texture_multisample, because it moves the code to compute the
physical size of multisampled surfaces out of renderbuffer-only code.
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Chad Versace <chad.versace@linux.intel.com>
Paul Berry [Tue, 8 Jan 2013 21:12:09 +0000 (13:12 -0800)]
intel: Add a force_y_tiling parameter to intel_miptree_create().
This allows intel_miptree_alloc_mcs() to force Y tiling for the MCS
buffer. Previously we accomplished this by the hack of passing
INTEL_MSAA_LAYOUT_CMS as the msaa_layout parameter, but that parameter
is going to be going away soon.
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Chad Versace <chad.versace@linux.intel.com>
Paul Berry [Tue, 8 Jan 2013 21:00:25 +0000 (13:00 -0800)]
intel: Move compute_msaa_layout earlier in file.
No functional change. This patch moves the compute_msaa_layout()
function earlier in intel_mipmap_tree.c so that it can be used by
other functions in that file.
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Chad Versace <chad.versace@linux.intel.com>
Vinson Lee [Wed, 9 Jan 2013 07:09:00 +0000 (08:09 +0100)]
r600g: Fix memory leak in r600_bytecode_add_vtx.
Fixes resource leak defect reported by Coverity.
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Marek Olšák [Wed, 9 Jan 2013 15:39:18 +0000 (16:39 +0100)]
r300g: optionally log MSAA resources to stderr
Set: RADEON_DEBUG=msaa
Marek Olšák [Wed, 9 Jan 2013 15:26:24 +0000 (16:26 +0100)]
r300g: fix the GPU name in the renderer string
Broken by
ca474f98f2cda5cb333e9f851.
Marek Olšák [Wed, 9 Jan 2013 10:34:33 +0000 (11:34 +0100)]
r300g: fix CS checker errors caused by emit_dsa_state
size is 10 on r500 and 8 on r300
Johannes Obermayr [Fri, 30 Nov 2012 00:44:56 +0000 (01:44 +0100)]
clover: Adapt libclc's INCLUDEDIR and LIBEXECDIR to make use of the new introduced libclc.pc.
Tom Stellard:
-Keep --with-libclc-path and mark it deprecated.
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Ian Romanick [Thu, 6 Dec 2012 22:57:01 +0000 (14:57 -0800)]
glsl: Don't add structure fields to the symbol table
I erroneously added this back in January 2011 in commit
88421589.
Looking at the commit message, I have no idea why I added it. It only
added non-array structure fields to the symbol table, so array structure
fields are treated correctly.
Fixes piglit tests structure-and-field-have-same-name.vert and
structure-and-field-have-same-name-nested.vert. It should also fix
WebGL conformance tests shader-with-non-reserved-words.
NOTE: This is a candidate for the stable release branches.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57622
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Kenneth Graunke [Tue, 8 Jan 2013 03:42:38 +0000 (19:42 -0800)]
i965/fs: Fix struct vs. class in acp_entry definitions.
Marek Olšák [Sat, 22 Dec 2012 18:33:47 +0000 (19:33 +0100)]
r600g: implement buffer copying using CP DMA for R7xx, Evergreen, Cayman
R6xx doesn't work - the issue seems to be with flushing (sometimes
the destination buffer contains garbage). There are no hangs, so we're good.
R7xx doesn't seem to have any alignment restriction despite our initial
thinking. Everything just works.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Marek Olšák [Tue, 8 Jan 2013 19:39:55 +0000 (20:39 +0100)]
st/mesa: fix possible MSVC build error v2
https://bugs.freedesktop.org/show_bug.cgi?id=59143
Using GLubyte as per Brian's suggestion.
Paul Berry [Wed, 19 Dec 2012 00:37:52 +0000 (16:37 -0800)]
glsl: Pack flat "varyings" of mixed types together.
This patch enhances the varying packing code so that flat varyings of
uint, int, and float types can be packed together.
We accomplish this in lower_packed_varyings.cpp by making the type of
all flat varyings ivec4, and then using information-preserving type
conversions (e.g. ir_unop_bitcast_f2i) to convert all other types to
ints.
The varying_matches::compute_packing_class() function is updated to
reflect the fact that varying packing no longer needs to segregate
varyings of different base types.
Fixes piglit test varying-packing-mixed-types.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
v2: Split lower_packed_varyings_visitor::bitwise_assign into
pack/unpack variants.
Paul Berry [Tue, 18 Dec 2012 23:24:39 +0000 (15:24 -0800)]
glsl: Prohibit structs and bools from being used as "varyings".
The GLSL 1.30 spec only allows vertex shader outputs and fragment
shader inputs ("varyings" in pre-GLSL-1.30 parlance) to be of type
int, uint, float, or vectors, matrices, or arrays thereof. Bools,
bvec's, and structs are prohibited. (Integral varyings were
prohibited prior to GLSL 1.30).
Previously, Mesa only performed this check on variables declared with
the "varying" keyword, and it always performed the check according to
the pre-GLSL-1.30 rules. As a result, bools and structs were allowed
to slip through, provided they were declared using the new in/out
syntax.
This patch modifies the error check so that it occurs after "varying"
is converted to "in/out", and corrects it to properly account for GLSL
version.
Fixes piglit tests:
in-bool-prohibited.frag
in-bvec2-prohibited.frag
in-bvec3-prohibited.frag
in-bvec4-prohibited.frag
in-struct-prohibited.frag
out-bool-prohibited.vert
out-bvec2-prohibited.vert
out-bvec3-prohibited.vert
out-bvec4-prohibited.vert
out-struct-prohibited.vert
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Paul Berry [Tue, 18 Dec 2012 22:49:34 +0000 (14:49 -0800)]
glsl: Plumb through is_parameter to apply_type_qualifier_to_variable()
This patch adds logic to allow the ast_to_hir function
apply_type_qualifier_to_variable() to tell whether it is acting on a
variable declaration or a function parameter. This will allow it to
correctly interpret the meaning of "out" and "in" keywords (which have
different meanings in those two contexts).
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Paul Berry [Mon, 17 Dec 2012 22:20:35 +0000 (14:20 -0800)]
glsl: Separate varying linking code to its own file.
linker.cpp is getting pretty big, and we're about to add even more
varying packing code, so split out the linker code that concerns
varyings to its own file.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Paul Berry [Mon, 17 Dec 2012 21:48:21 +0000 (13:48 -0800)]
mesa: Add ALIGN() macro to main/macros.h.
Previously this macro existed in 3 separate places, some inside the
intel driver and some outside of it. It makes more sense to have it
in main/macros.h
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Paul Berry [Tue, 8 Jan 2013 02:10:30 +0000 (18:10 -0800)]
glsl: Fix loop bounds detection.
When analyzing a loop where the loop condition is expressed in the
non-standard order (e.g. "4 > i" instead of "i < 4"), we were
reversing the condition incorrectly, leading to a loop bound that was
off by 1.
Fixes piglit tests {vs,fs}-loop-bounds-unrolled.shader_test.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Marek Olšák [Tue, 8 Jan 2013 15:38:10 +0000 (16:38 +0100)]
winsys/radeon: bump the size of relocation hashlist
This should reduce the number of hash collisions in ETQW.
Christoph Bumiller [Mon, 7 Jan 2013 14:46:31 +0000 (15:46 +0100)]
nvc0: catch too high GENERIC indices to prevent GRAPH traps
Christoph Bumiller [Mon, 7 Jan 2013 21:12:28 +0000 (22:12 +0100)]
nvc0: use correct resource target to select blit shader
Christoph Bumiller [Mon, 7 Jan 2013 19:18:06 +0000 (20:18 +0100)]
nvc0: add missing call to map edge flag in push_vbo
Note: this is a candidate for the 9.0 stable branch.
Christoph Bumiller [Mon, 7 Jan 2013 14:50:19 +0000 (15:50 +0100)]
nv50/ir: wrap assertion using typeid in #ifndef NDEBUG
Note: this is a candidate for the 9.0 stable branch.
Christoph Bumiller [Tue, 8 Jan 2013 12:46:24 +0000 (13:46 +0100)]
nvc0: fix out of bounds writes for unaligned sizes in push_data
Christoph Bumiller [Tue, 8 Jan 2013 11:35:25 +0000 (12:35 +0100)]
nouveau: increase max order of suballocated buffers by 1
This is really a hack to make TF2 (considerably, up to 20 -> 70 fps
at low res) faster.
Christoph Bumiller [Tue, 8 Jan 2013 15:13:11 +0000 (16:13 +0100)]
nouveau: improve buffer transfers
Save double memcpy on uploads to VRAM in most cases.
Properly handle FLUSH_EXPLICIT.
Reallocate on DISCARD_WHOLE_RESOURCE to avoid sync.
Marek Olšák [Tue, 8 Jan 2013 13:32:41 +0000 (14:32 +0100)]
r300g: fix assertion failure in emit_dsa_state
Broken by
8ed6b1400bc8a78f46340f41aaf2e88b24c23267.
Kenneth Graunke [Fri, 4 Jan 2013 15:53:12 +0000 (07:53 -0800)]
i965: Support GL_FIXED and packed vertex formats natively on Haswell+.
Haswell and later support the GL_FIXED and 2_10_10_10_rev vertex formats
natively, and don't need shader workarounds.
Reviewed-by: Eric Anholt <eric@anholt.net>
Kenneth Graunke [Fri, 4 Jan 2013 15:53:11 +0000 (07:53 -0800)]
i965: Add #defines for GL_FIXED vertex formats.
Reviewed-by: Eric Anholt <eric@anholt.net>
Kenneth Graunke [Fri, 4 Jan 2013 15:53:10 +0000 (07:53 -0800)]
i965: Add remaining #defines for packed vertex formats.
Reviewed-by: Eric Anholt <eric@anholt.net>
Kenneth Graunke [Fri, 4 Jan 2013 15:53:09 +0000 (07:53 -0800)]
i965: Use Haswell's sample_d_c for textureGrad with shadow samplers.
The new hardware actually just supports this now.
Reviewed-by: Eric Anholt <eric@anholt.net>
Kenneth Graunke [Mon, 7 Jan 2013 07:35:22 +0000 (23:35 -0800)]
i965/fs: Remove dead code from generate_uniform_pull_constant_load_gen7.
generate_uniform_pull_constant_load_gen7() is only called on Gen7+, so
the gen < 6 code is dead.
Reviewed-by: Eric Anholt <eric@anholt.net>
Alexander von Gluck IV [Sun, 6 Jan 2013 22:09:35 +0000 (16:09 -0600)]
mesa: Drop mmx optimizations on Haiku
* Prevents compatibility problems. As Haiku
doesn't use rtasm anymore, it's kind of
pointless.
Alexander von Gluck IV [Sun, 6 Jan 2013 22:06:37 +0000 (16:06 -0600)]
mesa: Don't use rtasm for Haiku swrast
* We have a symbol conflict as rtasm in
Mesa collides with rtasm in gallium.
* As us linking gallium and mesa together
is an edge case, lets just omit the rtasm
code from Mesa as we should be going
llvmpipe soon :)
Alex Deucher [Mon, 7 Jan 2013 20:21:46 +0000 (15:21 -0500)]
r600g: set the virtual address for the htile buffer
Fixes cayman and TN with htile enabled. Should fix:
https://bugs.freedesktop.org/show_bug.cgi?id=59089
https://bugs.freedesktop.org/show_bug.cgi?id=58667
Possibly others.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Jerome Glisse [Fri, 4 Jan 2013 21:34:52 +0000 (16:34 -0500)]
radeon/winsys: move radeon family/class identification to winsys
Upcoming async dma support rely on winsys knowing about GPU families.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Jerome Glisse [Fri, 4 Jan 2013 16:46:13 +0000 (11:46 -0500)]
r600g/radeon/winsys: indentation cleanup
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Marek Olšák [Sun, 6 Jan 2013 19:28:03 +0000 (20:28 +0100)]
r600g: flush FMASK and CMASK at the end of CS
Marek Olšák [Sat, 5 Jan 2013 05:21:49 +0000 (06:21 +0100)]
r300g: implement MSAA
This is not as optimized as r600g - the MSAA compression is missing,
so r300g needs a lot of bandwidth (more than r600g to do the same thing).
However, if the bandwidth is not an issue for you, you can enjoy this
unoptimized MSAA support.
The only other missing optimization for MSAA is the fast color clear.
MSAA is enabled on r500 only, because that's the only GPU family I tested.
That said, MSAA should work on r300 and r400 as well (but you must set
RADEON_MSAA=1 to allow it, then turn MSAA on in your app or set GALLIUM_MSAA=n,
n >= 2, n <= 6)
I will enable the support by default on r300-r400 once someone (other than me)
tests those chipsets with piglit.
The supported modes are 2x, 4x, 6x.
The supported MSAA formats are RGBA8, BGRA8, and RGBA16F (r500 only).
Those 3 formats are used for all GL internal formats.
Tested with piglit. (I have ported all MSAA tests to GL2.1)
Marek Olšák [Sun, 6 Jan 2013 00:47:24 +0000 (01:47 +0100)]
r300g: simplify DSA state, add ability to patch FG_ALPHA_FUNC while emitting
Preparation for MSAA and alpha-to-coverage.
Marek Olšák [Sat, 5 Jan 2013 23:31:55 +0000 (00:31 +0100)]
r300g/compiler: add shader emulation for the alpha_to_one state
Vinson Lee [Mon, 31 Dec 2012 23:19:43 +0000 (15:19 -0800)]
configure.ac: Remove space after indent -T flag.
Fixes this build error on platforms not using GNU indent.
indent: Command line: ``-T'' requires a parameter
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Ian Romanick [Fri, 30 Nov 2012 22:29:49 +0000 (14:29 -0800)]
intel: Fix copy-and-paste bug setting gl_constants::MaxSamples
gl_constants::MaxSamples is an integer, so setting it to 1.0 is just
silly.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
Reviewed-by: Eric Anholt <eric@anholt.net>
Ian Romanick [Mon, 3 Dec 2012 19:55:12 +0000 (11:55 -0800)]
mesa: Disallow R, RG, or RGB integer and unsigned formats in OpenGL ES 3.0
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Ian Romanick [Tue, 4 Dec 2012 18:36:10 +0000 (10:36 -0800)]
mesa: Disallow SNORM formats for renderbuffers in OpenGL ES
v2: Move {RED,RG,RGB,RGBA}_SNORM changes from the previous commit to
this commit. Based on suggestions from Ken.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Ian Romanick [Sat, 1 Dec 2012 19:06:48 +0000 (11:06 -0800)]
mesa: Disallow deprecated SNORM formats for renderbuffers
The OpenGL 3.2 core profile spec says:
"The following base internal formats from table 3.11 are
color-renderable: RED, RG, RGB, and RGBA. The sized internal formats
from table 3.12 that have a color-renderable base internal format
are also color-renderable. No other formats, including compressed
internal formats, are color-renderable."
The OpenGL 3.2 compatibility profile spec says (only ALPHA is added):
"The following base internal formats from table 3.16 are
color-renderable: ALPHA, RED, RG, RGB, and RGBA. The sized internal formats
from table 3.17 that have a color-renderable base internal format
are also color-renderable. No other formats, including compressed
internal formats, are color-renderable."
Table 3.12 in the core profile spec and table 3.17 in the compatibility
profile spec list SNORM formats as having a base internal format of RED,
RG, RGB, or RGBA. From this we infer that they should also be color
renderable.
The OpenGL ES 3.0 spec says:
"An internal format is color-renderable if it is one of the formats
from table 3.12 noted as color-renderable or if it is unsized format
RGBA or RGB. No other formats, including compressed internal
formats, are color-renderable."
In the OpenGL ES 3.0 spec, none of the SNORM formats have "color-
renderable" marked in table 3.12. The RGB I and UI formats also are not
color-renderable in ES3, but we'll save that change for another patch.
Both NVIDIA's closed-source driver (version 304.64) and AMD's
closed-source driver (Catalyst 12.6 on HD 3650) reject *all* SNORM
formats for renderbuffers in OpenGL 3.3 compatibility profiles.
v2: Move {RED,RG,RGB,RGBA}_SNORM changes from the this commit to the
next commit. Based on suggestions from Ken.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Brian Paul [Thu, 3 Jan 2013 15:04:50 +0000 (08:04 -0700)]
util: fix addressing bug in pipe_put_tile_z() for PIPE_FORMAT_Z32_FLOAT
The Z32 pixel is 4 bytes so multiply x by 4, not 2.
Note: This is a candidate for the stable branches.
Brian Paul [Thu, 3 Jan 2013 15:02:57 +0000 (08:02 -0700)]
util: add get/put_tile_z() support for PIPE_FORMAT_Z32_FLOAT_S8X24_UINT
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=58972
Note: This is a candidate for the stable branches.
Brian Paul [Wed, 2 Jan 2013 20:46:20 +0000 (13:46 -0700)]
gallivm: support more immediates in lp_build_tgsi_info()
Bump limit from 32 to 128.
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=58545
Brian Paul [Fri, 4 Jan 2013 00:31:22 +0000 (17:31 -0700)]
xlib: allow GLX_DONT_CARE for glXChooseFBConfig() attribute values
Fixes piglit glx-dont-care-mask test.
Note: This is a candidate for the stable branches.
Reviewed-by: Chad Versace <chad.versace@linux.intel.com>
Brian Paul [Fri, 4 Jan 2013 00:30:34 +0000 (17:30 -0700)]
st/glx: allow GLX_DONT_CARE for glXChooseFBConfig() attribute values
Fixes piglit glx-dont-care-mask test.
Note: This is a candidate for the stable branches.
Reviewed-by: Chad Versace <chad.versace@linux.intel.com>
Tom Stellard [Fri, 4 Jan 2013 15:38:37 +0000 (15:38 +0000)]
radeon/llvm: Remove backend code from Mesa
This code now lives in an external tree.
For the next Mesa release fetch the code from the master branch
of this LLVM repo:
http://cgit.freedesktop.org/~tstellar/llvm/
For all subsequent Mesa releases, fetch the code from the official LLVM
project:
www.llvm.org
Johannes Obermayr [Thu, 20 Dec 2012 19:56:17 +0000 (20:56 +0100)]
Support LLVM >= 3.2 on radeonsi and opencl.
Tom Stellard:
- Backend now has same name for all LLVM versions
- Add missing LLVM_VERSION_INT definition
Tom Stellard [Fri, 4 Jan 2013 15:47:53 +0000 (15:47 +0000)]
clover: Fix build after the addition of enum pipe_flush_flags
Broken since commit
598cc1f74d7ae924e84dee801b456ab7b0b22f84
Marek Olšák [Wed, 2 Jan 2013 19:28:10 +0000 (20:28 +0100)]
r300g: don't check for vertex and index buffer bind flags
Marek Olšák [Fri, 4 Jan 2013 19:07:06 +0000 (20:07 +0100)]
r300g/swtcl: use memcpy to emit indices
Marek Olšák [Fri, 4 Jan 2013 17:00:46 +0000 (18:00 +0100)]
r300g/swtcl: simplify vertex uploading
- skip the vertex buffer reallocation in flush and just use
the unsynchronized flag to get new memory.
- remove the cruft needed to get around the issues with the vertex buffer
reallocation in flush
- use pb_buffer instead of pipe_resource
Marek Olšák [Fri, 4 Jan 2013 17:27:21 +0000 (18:27 +0100)]
r300g/swtcl: fix crash when setting vertex buffers
Broken by
e73bf3b805de78299f1a652668ba4e6eab9bac94.
Marek Olšák [Wed, 2 Jan 2013 19:23:57 +0000 (20:23 +0100)]
r300g: don't set PIPE_BIND flags for internal textures
Paul Berry [Sat, 29 Dec 2012 19:31:37 +0000 (11:31 -0800)]
i965: Fix glCompressedTexSubImage2D offsets for ETC textures.
This patch fixes intel_miptree_unmap_etc() (which decompresses ETC
textures to linear) to pay attention to map->x and map->y when writing
to the destination image. Previously these values were ignored,
causing the xoffset and yoffset parameters passed to
glCompressedTexSubImage2D() to be ignored.
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Kristian Høgsberg [Fri, 14 Dec 2012 04:32:51 +0000 (23:32 -0500)]
egl/wayland: Remove kooky flush code
We used to have to jump through hoops to call glFlush at swap buffer time,
but the flush extension made that unnecessary a long time ago.
Kristian Høgsberg [Fri, 14 Dec 2012 04:21:46 +0000 (23:21 -0500)]
egl/wayland: Remove confusing comment about front buffer rendering
Kristian Høgsberg [Fri, 14 Dec 2012 04:20:16 +0000 (23:20 -0500)]
egl_dri2: Remove unused struct dri2_egl_buffer from header file
Kristian Høgsberg [Thu, 13 Dec 2012 20:59:24 +0000 (15:59 -0500)]
egl: Add extension infrastructure for EGL_EXT_buffer_age
Kristian Høgsberg [Fri, 14 Dec 2012 04:44:09 +0000 (23:44 -0500)]
egl: Update to revision 19987 of eglext.h
This pulls in EGL_EXT_buffer_age.
Brian Paul [Fri, 4 Jan 2013 15:21:12 +0000 (08:21 -0700)]
util: move var declaration before loop to fix MSVC error
Marek Olšák [Thu, 20 Dec 2012 02:45:49 +0000 (03:45 +0100)]
r600g: implement 3D transfers
That means we can map and read multiple slices with one transfer_map call.
Marek Olšák [Thu, 20 Dec 2012 16:00:06 +0000 (17:00 +0100)]
st/mesa: fix assertion failures with
2101010 vertex formats
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Thu, 20 Dec 2012 15:40:33 +0000 (16:40 +0100)]
st/mesa: accelerate CopyTexSubImage for 1D array textures
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Thu, 20 Dec 2012 14:15:15 +0000 (15:15 +0100)]
st/mesa: fix CopyTexSubImage fallback for 1D array textures
- We should use a 3D transfer of size Width x 1 x NumLayers.
- We should use layer_stride instead of stride.
(even though they are likely to be equal with 1D array textures)
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Thu, 20 Dec 2012 02:54:33 +0000 (03:54 +0100)]
st/mesa: fix GetTexImage for compressed 2D array textures
This uses a 3D blit to decompress the texture and then a 3D transfer
to read it.
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Thu, 20 Dec 2012 01:33:45 +0000 (02:33 +0100)]
gallium/util: remove unused helper util_create_rgba_texture
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Thu, 20 Dec 2012 01:09:56 +0000 (02:09 +0100)]
st/mesa: try to find the format matching format+type in decompressed_with_blit
There was the fast path based on _mesa_format_matches_format_and_type
for GetTexImage, but it never worked, because the Mesa format we were testing
there was always compressed. Further testing showed that the fast path
had been completely broken.
In this commit, the somewhat limited helper util_create_rgba_texture is
no longer used and instead, custom code for the texture creation is added,
which tries to find the best matching RGBA8 format, so that we can hit
the fast path *always* if the read format is a variant of RGBA8 and supported
by the driver.
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Thu, 20 Dec 2012 00:41:57 +0000 (01:41 +0100)]
st/mesa: fix GetTexImage for compressed cubemaps
I'll deal with 2D arrays later.
NOTE: This is a candidate for the stable branches.
Marek Olšák [Thu, 20 Dec 2012 02:43:57 +0000 (03:43 +0100)]
gallium/u_blitter: implement 3D blitting
Scaling and flipping in the Z direction isn't allowed yet.
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Thu, 20 Dec 2012 12:08:00 +0000 (13:08 +0100)]
gallium/u_blitter: fix blitting TEXTURE_CUBE_ARRAY with a non-zero cube index
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Thu, 20 Dec 2012 02:42:11 +0000 (03:42 +0100)]
gallium/u_blitter: minor simplification
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Wed, 19 Dec 2012 22:06:51 +0000 (23:06 +0100)]
gallium/u_blitter: unify some parameters into a dstbox parameter in blit_generic
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Wed, 19 Dec 2012 21:25:22 +0000 (22:25 +0100)]
gallium/u_blitter: remove useless parameter from blitter_default_dst_texture
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Thu, 20 Dec 2012 14:07:35 +0000 (15:07 +0100)]
gallium/util: complete implementation of util_dump_transfer
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Thu, 20 Dec 2012 02:57:39 +0000 (03:57 +0100)]
mesa: allow TEXTURE_CUBE_MAP_ARRAY in GetTexImage
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Fri, 21 Dec 2012 16:15:56 +0000 (17:15 +0100)]
gallium/radeon: send the END_OF_FRAME flag to the DRM
Marek Olšák [Fri, 21 Dec 2012 16:03:22 +0000 (17:03 +0100)]
gallium: extend pipe_context::flush for it to accept an END_OF_FRAME flag
Usage with pipe_context:
pipe->flush(pipe, NULL, PIPE_FLUSH_END_OF_FRAME);
Usage with st_context_iface:
st->flush(st, ST_FLUSH_END_OF_FRAME, NULL);
The flag is only a hint for drivers. Radeon will use it for buffer eviction
heuristics in the kernel (e.g. for queries like how many frames have passed
since a buffer was used).
The flag is currently only generated by st/dri on SwapBuffers.
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
Marek Olšák [Fri, 4 Jan 2013 11:40:33 +0000 (12:40 +0100)]
radeonsi: fix int->bool conversion in fence_signalled
Marek Olšák [Fri, 4 Jan 2013 11:40:04 +0000 (12:40 +0100)]
r600g: fix int->bool conversion in fence_signalled
NOTE: This is a candidate for the stable branches.
Paul Berry [Thu, 3 Jan 2013 19:51:38 +0000 (11:51 -0800)]
Add new .gitignore entries for Automake 1.13 tests
Automake 1.13 creates a bunch of new build artefacts:
- bin/test-driver, a script for running tests.
- *.trs files for every "make check" test result.
- *.log files containing the output of every test run by "make check".
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Kenneth Graunke [Fri, 28 Dec 2012 20:45:14 +0000 (12:45 -0800)]
i965: Replace structs with bit-shifting for Gen7 SURFACE_STATE entries.
Every generation except Gen7 creates SURFACE_STATE entries via a
uint32_t array. Only Gen7 uses the older bitfield structure, which we
moved away from because it was less efficient. Convert it for
consistency.
This reduces the compiled size of gen7_wm_surface_state.o by 2.86% in a
release build.
v2: Fix accidental use of BRW_SURFACE_WIDTH/HEIGHT in brw_state_dump.c;
switch back to gen7_set_surface_mcs_info setting surf[6] directly
(both per Eric's review comments).
Acked-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
smoki [Thu, 3 Jan 2013 17:57:40 +0000 (18:57 +0100)]
radeon/r200: Fix tcl culling
Should fix:
https://bugs.freedesktop.org/show_bug.cgi?id=57842
Jonas Ådahl [Tue, 25 Dec 2012 12:01:08 +0000 (13:01 +0100)]
wayland: Don't cancel a roundtrip when any event is received
Since wl_display_dispatch_queue() returns the number of processed events
or -1 on error, only cancel the roundtrip if an -1 is returned.
This also fixes a potential memory corruption bug happening when the
roundtrip does an early return and the callback later writes to the then
out of scope stack allocated `done' parameter.
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Vinson Lee [Wed, 2 Jan 2013 01:02:38 +0000 (17:02 -0800)]
i965: Add break statement at end of BRW_OPCODE_CONTINUE case.
Fixes missing break in switch defect reported by Coverity.
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Paul Berry <stereotype441@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Chad Versace [Thu, 20 Dec 2012 22:16:50 +0000 (14:16 -0800)]
egl/android: Fix build for Jelly Bean (v2)
In Jelly Bean, the interface to ANativeWindow changed. The change included
adding a new parameter the queueBuffer and dequeueBuffer methods,
removing the lockBuffer method, and requiring libsync.
v2:
- s/fence_fd == -1/fence_fd != -1/
- Fix leak. Close the fence_fd.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Signed-off-by: Chad Versace <chad.versace@linux.intel.com>
Chad Versace [Thu, 20 Dec 2012 22:14:51 +0000 (14:14 -0800)]
android: Define Make variables for Android version
Define the following Make variables:
MESA_ANDROID_MAJOR_VERSION
MESA_ANDROID_MINOR_VERSION
MESA_ANDROID_VERSION
These variable will allow us to make version-dependent decisions on
library dependencies. In particular, building Mesa against JellyBean will
require libsync.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Signed-off-by: Chad Versace <chad.versace@linux.intel.com>
Matt Turner [Mon, 31 Dec 2012 21:41:57 +0000 (13:41 -0800)]
mesa: Add missing ASSERT_OUTSIDE_BEGIN_END to GetSamplerParameter*
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>