mesa.git
8 years agogenxml: Put append counter fields before MCS in RENDER_SURFACE_STATE on gen7
Jason Ekstrand [Tue, 7 Jun 2016 18:37:46 +0000 (11:37 -0700)]
genxml: Put append counter fields before MCS in RENDER_SURFACE_STATE on gen7

The pack header generation scripts can't handle the case where you have
two addresses in the same dword; they just take whatever is the last one.
This meant that the MCS address wasn't properly getting handled.  Since we
don't care about append counters, we can just re-arrange the XML for now.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoanv,isl: Lower storage image formats in anv
Jason Ekstrand [Thu, 9 Jun 2016 20:56:16 +0000 (13:56 -0700)]
anv,isl: Lower storage image formats in anv

ISL was being a bit too clever for its own good and lowering the format for
us.  This is all well and good *if* we always want to lower it.  However,
the GL driver selectively lowers the format depending on whether the
surface is write-only or not.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Allow for full 31-bit buffer texture sizes
Jason Ekstrand [Fri, 10 Jun 2016 17:45:43 +0000 (10:45 -0700)]
isl/state: Allow for full 31-bit buffer texture sizes

Ivy Bridge and above can handle up to 2^31 elements for RAW buffer
surfaces.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Don't use designated initializers for buffer surface state
Jason Ekstrand [Sat, 11 Jun 2016 01:28:49 +0000 (18:28 -0700)]
isl/state: Don't use designated initializers for buffer surface state

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Add assertions for buffer surface restrictions
Jason Ekstrand [Fri, 10 Jun 2016 01:48:07 +0000 (18:48 -0700)]
isl/state: Add assertions for buffer surface restrictions

Acked-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Don't set SurfacePitch for gen9 1-D textures
Jason Ekstrand [Thu, 9 Jun 2016 22:08:13 +0000 (15:08 -0700)]
isl/state: Don't set SurfacePitch for gen9 1-D textures

This field is ignored by the hardware in this case and, on very large 1-D
textures, it can end up being larger than the maximum allowed value.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Use TILEWALK_XMAJOR for linear surfaces on gen7
Jason Ekstrand [Tue, 7 Jun 2016 21:33:40 +0000 (14:33 -0700)]
isl/state: Use TILEWALK_XMAJOR for linear surfaces on gen7

This matches better what happens on gen8 where the "Tiled Surface" and
"Tile Walke" bits are combined into a single two-bit value.  This is also
more consistent with what the GL driver does.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Emit no-op mip tail setup on SKL
Jason Ekstrand [Tue, 7 Jun 2016 01:22:21 +0000 (18:22 -0700)]
isl/state: Emit no-op mip tail setup on SKL

This hasn't ever been a problem in the past but it is recommended by the
hardware docs.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Only set cube face enables if usage includes CUBE_BIT
Jason Ekstrand [Tue, 7 Jun 2016 01:21:17 +0000 (18:21 -0700)]
isl/state: Only set cube face enables if usage includes CUBE_BIT

It seems safe to set it all the time, but this reduces the diff between
the way i965 does it and what ISL does.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Use the layout for computing qpitch rather than dimensions
Jason Ekstrand [Sun, 5 Jun 2016 04:22:21 +0000 (21:22 -0700)]
isl/state: Use the layout for computing qpitch rather than dimensions

For depth/stencil 1-D textures on SKL, we want them layed out in the old
format that has been used since gen4.  In order for the surface state
fill-out code to handle, this it needs to distinguish based on layout
rather than just dimensionality.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Set the IntegerSurfaceFormat bit on Haswell
Jason Ekstrand [Tue, 7 Jun 2016 15:05:45 +0000 (08:05 -0700)]
isl/state: Set the IntegerSurfaceFormat bit on Haswell

This fixes 688 Vulkan CTS tests on Haswell.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/format: Mark R9G9B9E5 as containing 9-bit unsigned float channels
Jason Ekstrand [Tue, 7 Jun 2016 21:40:47 +0000 (14:40 -0700)]
isl/format: Mark R9G9B9E5 as containing 9-bit unsigned float channels

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Don't set RenderTargetViewExtent for texture surfaces
Jason Ekstrand [Fri, 17 Jun 2016 23:40:24 +0000 (16:40 -0700)]
isl/state: Don't set RenderTargetViewExtent for texture surfaces

The docs specify that this only matters for render targets and surfaces
used with typed dataport messages.  On some platforms (gen4-6) the Depth
field has more bits than RenderTargetViewExtent so we can have textures
with more levels than we can render to.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Set SurfaceArray based on the surface dimension
Jason Ekstrand [Tue, 7 Jun 2016 22:30:00 +0000 (15:30 -0700)]
isl/state: Set SurfaceArray based on the surface dimension

According to the PRM, you can't set SurfaceArray for 3D or buffer textures.
There doesn't seem to be a good reason not to set it when we can.  On the
other hand, if we don't set it we can end up getting strange results for
1-layer array textures such as textureSize() returning the wrong results.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Don't force-disable L2 bypass for everything
Jason Ekstrand [Sun, 5 Jun 2016 03:48:55 +0000 (20:48 -0700)]
isl/state: Don't force-disable L2 bypass for everything

We already set the bit in the few cases where it's required by the docs so
there's no need to set it all the time.  This has no noticable perf impact
for Dota 2 on Vulkan with the time demo I have.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Refactor the setup of clear colors
Jason Ekstrand [Fri, 3 Jun 2016 02:02:23 +0000 (19:02 -0700)]
isl/state: Refactor the setup of clear colors

This commit switches clear colors to use #if's instead of a C if.  This
lets us properly handle SNB where the clear color field doesn't exist.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Refactor the per-gen isl_to_gen_h/valign tables
Jason Ekstrand [Fri, 3 Jun 2016 02:00:10 +0000 (19:00 -0700)]
isl/state: Refactor the per-gen isl_to_gen_h/valign tables

This moves the #if's around so that halign and valign have different sets
of #if conditions.  This also prepares us for SNB because isl_to_gen_halign
is not defined at all on gen6.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Return an extent3d from the halign/valign helper
Jason Ekstrand [Fri, 3 Jun 2016 01:40:07 +0000 (18:40 -0700)]
isl/state: Return an extent3d from the halign/valign helper

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Put pitch calculations together
Jason Ekstrand [Wed, 8 Jun 2016 00:01:56 +0000 (17:01 -0700)]
isl/state: Put pitch calculations together

This is purely cosmetic, but it makes things look a bit more readable.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Put all dimension setup together and towards the top
Jason Ekstrand [Tue, 7 Jun 2016 23:58:54 +0000 (16:58 -0700)]
isl/state: Put all dimension setup together and towards the top

This is purely cosmetic, but it makes things look a bit more readable.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Put surface format setup at the top
Jason Ekstrand [Wed, 8 Jun 2016 00:14:39 +0000 (17:14 -0700)]
isl/state: Put surface format setup at the top

This is purely cosmetic, but it makes things look a bit more readable.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Remove some unused fields
Jason Ekstrand [Tue, 7 Jun 2016 23:55:21 +0000 (16:55 -0700)]
isl/state: Remove some unused fields

They're already zero-initialized and we have no plans of doing anything
more interesting with them.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoisl/state: Don't use designated initializers for the surface state
Jason Ekstrand [Tue, 7 Jun 2016 23:53:19 +0000 (16:53 -0700)]
isl/state: Don't use designated initializers for the surface state

While designated initializers are nice, they also force us to put some
things in the initializer and some things later.  Surface state setup is
complicated enough that this really hurts readability in the long run.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agogenxml/gen8,9: Prefix the multisample format enum with MSFMT
Jason Ekstrand [Fri, 3 Jun 2016 01:43:59 +0000 (18:43 -0700)]
genxml/gen8,9: Prefix the multisample format enum with MSFMT

This is what gen7 does and it's nice to have a prefix

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoi965/blorp: Only set src_z for gen8+ 3D textures
Jason Ekstrand [Fri, 10 Jun 2016 21:44:32 +0000 (14:44 -0700)]
i965/blorp: Only set src_z for gen8+ 3D textures

Otherwise, we end up with a bogus value in the third component.  On gen6-7
where we always use 2D textures, this can cause problems if the
SurfaceArray bit is set in the SURFACE_STATE.

Acked-by: Chad Versace <chad.versace@intel.com>
8 years agoi965/gen7,8: Set SURFACE_IS_ARRAY for all non-3D texture types
Jason Ekstrand [Sat, 11 Jun 2016 14:51:30 +0000 (07:51 -0700)]
i965/gen7,8: Set SURFACE_IS_ARRAY for all non-3D texture types

There's no real reason why we shouldn't set this bit.  It does affect how
the sampler operates a bit but since you can have a 2D non-array view of a
2D_ARRAY texture that distinction is very weak.  Also, this is what ISL
will do and we would like this change to be isolated from using ISL.

Reviewed-by: Chad Versace <chad.versace@intel.com>
8 years agoi965/gen4: Subtract 1 from buffer sizes
Jason Ekstrand [Sat, 11 Jun 2016 04:11:02 +0000 (21:11 -0700)]
i965/gen4: Subtract 1 from buffer sizes

The PRM states that the values put in Width, Height, and Depth should be
various bits from the value size - 1.  We seem to have done this wrong
more-or-less from the start.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Cc: "11.1 11.2 12.0" <mesa-stable@lists.freedesktop.org>
8 years agoi965: Remove fake W-tiled render target support
Jason Ekstrand [Thu, 9 Jun 2016 22:52:52 +0000 (15:52 -0700)]
i965: Remove fake W-tiled render target support

This hasn't been used since 1cfb4bc890b8 where we deleted the meta stencil
blit path.

Reviewed-by: Chad Versace <chad.versace@intel.com>
8 years agoi965/fs: Use a default Y coordinate of 0 for TXF on gen9+
Jason Ekstrand [Tue, 7 Jun 2016 02:15:39 +0000 (19:15 -0700)]
i965/fs: Use a default Y coordinate of 0 for TXF on gen9+

Previously, we were incrementing length but not actually putting anything
in the Y coordinate.  This meant that 1-D TXF operations had a garbage
array index.  If the surface is emitted as 1-D non-array, the coordinate
gets discarded and it works fine.  If it happens to be bound as an array
surface, it may count as an out-of-bounds array access and you get zero.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Cc: "11.1 11.2 12.0" <mesa-stable@lists.freedesktop.org>
8 years agoi965/gen8: Use the qpitch from the aux_mt for AUX_QPITCH
Jason Ekstrand [Sat, 4 Jun 2016 21:32:37 +0000 (14:32 -0700)]
i965/gen8: Use the qpitch from the aux_mt for AUX_QPITCH

Reviewed-by: Chad Versace <chad.versace@intel.com>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Cc: "11.1 11.2 12.0" <mesa-stable@lists.freedesktop.org>
8 years agoi965/blorp/gen8: Use the correct max level and layer in emit_surface_states
Jason Ekstrand [Sat, 4 Jun 2016 06:25:19 +0000 (23:25 -0700)]
i965/blorp/gen8: Use the correct max level and layer in emit_surface_states

We were adding in the base which is wrong because the values given in the
miptree are relative to zero and not the base layer/level.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Cc: "11.1 11.2 12.0" <mesa-stable@lists.freedesktop.org>
8 years agoi965: Drop the maximum 3D texture size to 512 on Sandy Bridge
Jason Ekstrand [Thu, 9 Jun 2016 21:57:33 +0000 (14:57 -0700)]
i965: Drop the maximum 3D texture size to 512 on Sandy Bridge

The RenderTargetViewExtent field of RENDER_SURFACE_STATE is supposed to be
set to the depth of a 3-D texture when rendering.  Unfortunatley, that
field is only 9 bits on Sandy Bridge and prior so we can't actually bind
a 3-D texturing for rendering if it has depth > 512.  On Ivy Bridge, this
field was bumpped to 11 bits so we can go all the way up to 2048.  On Iron
Lake and prior, we don't support layered rendering and we use OffsetX/Y
hacks to render to particular layers so 2048 is ok there too.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Cc: "11.1 11.2 12.0" <mesa-stable@lists.freedesktop.org>
8 years agoi965/gen4-6: Handle gl_texture_object::BaseLevel and MinLayer correctly
Jason Ekstrand [Wed, 22 Jun 2016 18:11:29 +0000 (11:11 -0700)]
i965/gen4-6: Handle gl_texture_object::BaseLevel and MinLayer correctly

This is basically a direct translation of what we do for gen7.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=83036
Cc: "11.1 11.2 12.0" <mesa-stable@lists.freedesktop.org>
8 years agoi965/gen4: Pull texture formats from the texture object not the miptree
Jason Ekstrand [Wed, 22 Jun 2016 04:58:23 +0000 (21:58 -0700)]
i965/gen4: Pull texture formats from the texture object not the miptree

This makes texture views sort-of work.  It doesn't add full texture view
support for gen4-5 but it is enough to fix the GL_ARB_copy_image formats
piglit test on Iron Lake.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=83036
Cc: "11.1 11.2 12.0" <mesa-stable@lists.freedesktop.org>
8 years agoi965: Fix point size with tessellation/geometry shaders in GLES.
Kenneth Graunke [Thu, 2 Jun 2016 00:32:55 +0000 (17:32 -0700)]
i965: Fix point size with tessellation/geometry shaders in GLES.

Our previous code worked for desktop GL, and ES without geometry or
tessellation shaders.  But those features require fancier point size
handling.  Fortunately, we can use one rule for all APIs.

Fixes a number of dEQP tests with EXT_tessellation_shader enabled:
dEQP-GLES31.functional.tessellation_geometry_interaction.point_size.*

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
8 years ago.mailmap: fix my main address
Marek Olšák [Wed, 22 Jun 2016 12:45:04 +0000 (14:45 +0200)]
.mailmap: fix my main address

8 years agoi965: move vs outputs written into a helper
Timothy Arceri [Tue, 26 Apr 2016 09:52:45 +0000 (19:52 +1000)]
i965: move vs outputs written into a helper

We will reuse this for fs key generation for the on disk shader
cache.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
8 years agost/mesa: use a single memcpy in st_ReadPixels when possible
Nicolai Hähnle [Tue, 14 Jun 2016 18:03:53 +0000 (20:03 +0200)]
st/mesa: use a single memcpy in st_ReadPixels when possible

This avoids costly address recomputations, function overhead, and may trigger
large copy optimizations.

Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agoglsl: only match gl_FragData and not gl_SecondaryFragDataEXT
Ilia Mirkin [Tue, 21 Jun 2016 20:16:17 +0000 (16:16 -0400)]
glsl: only match gl_FragData and not gl_SecondaryFragDataEXT

There's special logic around finding gl_FragData. It latches onto any
array with FRAG_RESULT_DATA0. However gl_SecondaryFragDataEXT[], added
by GL_EXT_blend_func_extended, fits those parameters as well. The real
frag data array should have index 0 though, so we can use that to
distinguish them.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96617
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "11.1 11.2 12.0" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
8 years agonv50,nvc0: fix start_instance in manual push path
Ilia Mirkin [Sun, 19 Jun 2016 01:54:37 +0000 (21:54 -0400)]
nv50,nvc0: fix start_instance in manual push path

The start instance is applied as an offset into the buffer directly,
ignoring the divisor, not as an instance id offset that respects the
divisor.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "11.2 12.0" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
8 years agotranslate: fix start_instance parameter in sse version
Ilia Mirkin [Sun, 19 Jun 2016 04:43:06 +0000 (00:43 -0400)]
translate: fix start_instance parameter in sse version

The generic version gets this right already, but this was using an
incorrect formula in SSE.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "11.2 12.0" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
8 years agoanv/cmd: Dirty descriptor sets when a new pipeline is bound
Jason Ekstrand [Tue, 21 Jun 2016 22:32:09 +0000 (15:32 -0700)]
anv/cmd: Dirty descriptor sets when a new pipeline is bound

Ever since c2581a9375ea, the binding table layout has depended on the
pipeline.  This means that whenever we change pipelines we also need to
re-emit binding tables for the new layout.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoanv/cmd: Move emit_descriptor_pointers to genX_cmd_buffer.c
Jason Ekstrand [Tue, 21 Jun 2016 22:31:14 +0000 (15:31 -0700)]
anv/cmd: Move emit_descriptor_pointers to genX_cmd_buffer.c

It's tiny and fully generic so there's really no reason for it to be in a
gen7-specific file.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoanv/cmd: Move flush_descriptor_sets to anv_cmd_buffer.c
Jason Ekstrand [Tue, 21 Jun 2016 22:28:15 +0000 (15:28 -0700)]
anv/cmd: Move flush_descriptor_sets to anv_cmd_buffer.c

There's no good reason for recompiling it

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agospirv: Use the system value version of gl_FrontFace
Jason Ekstrand [Tue, 21 Jun 2016 06:41:11 +0000 (23:41 -0700)]
spirv: Use the system value version of gl_FrontFace

SPIR-V treats it as an input but NIR wants the system value.  This
shouldn't have been too much of a surprise given that we have to do the
same conversion in the GLSL IR to NIR pass.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoi965: Reorganize prog_data->total_scratch code a bit.
Kenneth Graunke [Tue, 14 Jun 2016 06:09:31 +0000 (23:09 -0700)]
i965: Reorganize prog_data->total_scratch code a bit.

Cc: "12.0" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
8 years agoradeonsi: add a debug flag for unsafe math LLVM optimizations
Marek Olšák [Fri, 10 Jun 2016 20:43:29 +0000 (22:43 +0200)]
radeonsi: add a debug flag for unsafe math LLVM optimizations

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: use u_blitter for mipmap generation
Marek Olšák [Fri, 10 Jun 2016 01:03:11 +0000 (03:03 +0200)]
radeonsi: use u_blitter for mipmap generation

This reduces time spend in glGenerateMipmap by a half.

v2: don't decompress the levels to be overwritten

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium/u_blitter: implement mipmap generation
Marek Olšák [Fri, 10 Jun 2016 01:02:24 +0000 (03:02 +0200)]
gallium/u_blitter: implement mipmap generation

for pipe_context::generate_mipmap

first move some of the blit code from util_blitter_blit_generic
to a separate function, then use it from util_blitter_generate_mipmap

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agost/mesa: cache staging texture for glReadPixels
Nicolai Hähnle [Wed, 8 Jun 2016 11:24:14 +0000 (13:24 +0200)]
st/mesa: cache staging texture for glReadPixels

v2: add ST_DEBUG flag for disabling (suggested by Ilia)

Reviewed-by: Marek Olšák <marek.olsak@amd.com> (v1)
8 years agost/mesa: invalidate readpixels cache
Nicolai Hähnle [Thu, 9 Jun 2016 10:16:26 +0000 (12:16 +0200)]
st/mesa: invalidate readpixels cache

Whenever a draw happens or some other function call might change the result
of future glReadPixels calls, we must invalidate the cache.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agost/mesa: add readpix_cache structure
Nicolai Hähnle [Wed, 8 Jun 2016 11:22:52 +0000 (13:22 +0200)]
st/mesa: add readpix_cache structure

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agost/mesa: move ReadPixels blit into a separate function
Nicolai Hähnle [Wed, 8 Jun 2016 11:01:05 +0000 (13:01 +0200)]
st/mesa: move ReadPixels blit into a separate function

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agost/mesa: flush bitmap cache before CopyImageSubData
Nicolai Hähnle [Fri, 10 Jun 2016 13:59:58 +0000 (15:59 +0200)]
st/mesa: flush bitmap cache before CopyImageSubData

Found by inspection.

Cc: 11.2 12.0 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agost/mesa: flush bitmap cache before texture functions
Nicolai Hähnle [Thu, 9 Jun 2016 10:22:31 +0000 (12:22 +0200)]
st/mesa: flush bitmap cache before texture functions

As far as I can tell, a sequence of glBitmap followed by texture functions
that refer to a texture bound as the framebuffer is well within what should
be allowed.

Found by inspection.

Cc: 11.2 12.0 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agost/mesa: flush bitmap cache before compute dispatch
Nicolai Hähnle [Thu, 9 Jun 2016 10:12:34 +0000 (12:12 +0200)]
st/mesa: flush bitmap cache before compute dispatch

In the unlikely case that a program uses glBitmap to render to a framebuffer
whose texture is bound in a compute shader.

Found by inspection.

Cc: 11.2 12.0 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agoi965: get PrimitiveMode from the program rather than the shader struct
Timothy Arceri [Mon, 20 Jun 2016 07:40:23 +0000 (17:40 +1000)]
i965: get PrimitiveMode from the program rather than the shader struct

This is more consistent with what we do elsewhere and will allow
us to only cache one of the values in the shader cache.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
8 years agoclover: Fix build against clang SVN >= r273191
Vedran Miletić [Mon, 20 Jun 2016 23:17:49 +0000 (01:17 +0200)]
clover: Fix build against clang SVN >= r273191

setLangDefaults() now requires PreprocessorOptions as an argument.

Reviewed-and-Tested-by: Michel Dänzer <michel.daenzer@amd.com>
8 years agoi965: Fix multiplication of immediates on Cherryview/Broxton.
Kenneth Graunke [Wed, 8 Jun 2016 23:09:02 +0000 (16:09 -0700)]
i965: Fix multiplication of immediates on Cherryview/Broxton.

Cherryview and Broxton don't support DW x DW multiplication.  We have
piles of code to handle this, but apparently weren't retyping in the
immediate case.

For example,
tests/spec/arb_tessellation_shader/execution/dvec3-vs-tcs-tes
makes the simulator angry about instructions such as:

   mul(8) r18<1>:D r10.0<8;8,1>:D 0x00000003:D

Just retype to W or UW.  It should be safe on all platforms.

Cc: "12.0" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95462
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
8 years agoanv: Add proper support for depth clamping
Jason Ekstrand [Tue, 14 Jun 2016 15:40:49 +0000 (08:40 -0700)]
anv: Add proper support for depth clamping

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoanv/cmd_buffer: Split emit_viewport in two
Jason Ekstrand [Tue, 14 Jun 2016 15:15:34 +0000 (08:15 -0700)]
anv/cmd_buffer: Split emit_viewport in two

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoanv/cmd_buffer: Set depth/stencil extent based on the image
Jason Ekstrand [Tue, 14 Jun 2016 00:09:37 +0000 (17:09 -0700)]
anv/cmd_buffer: Set depth/stencil extent based on the image

It used to be based on the framebuffer which isn't quite right.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoanv/cmd_buffer: Don't crash if push constants are provided for missing stages
Jason Ekstrand [Wed, 15 Jun 2016 21:30:33 +0000 (14:30 -0700)]
anv/cmd_buffer: Don't crash if push constants are provided for missing stages

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agoanv/pipeline: Do invariance propagation on SPIR-V shaders
Jason Ekstrand [Thu, 16 Jun 2016 17:57:39 +0000 (10:57 -0700)]
anv/pipeline: Do invariance propagation on SPIR-V shaders

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agonir/alu_to_scalar: Respect the exact ALU operation qualifier
Jason Ekstrand [Mon, 13 Jun 2016 21:41:05 +0000 (14:41 -0700)]
nir/alu_to_scalar: Respect the exact ALU operation qualifier

Just setting builder->exact isn't sufficient because that only applies to
instructions that are built with the builder but instructions created
manually and only inserted using the builder are left alone.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agonir: Add a pass for propagating invariant decorations
Jason Ekstrand [Mon, 13 Jun 2016 19:47:19 +0000 (12:47 -0700)]
nir: Add a pass for propagating invariant decorations

This pass is similar to propagate_invariance in the GLSL compiler.  The
real "output" of this pass is that any algebraic operations which are
eventually consumed by an invariant variable get marked as "exact".

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agonir/algebraic: Remove imprecise flog2 optimizations
Jason Ekstrand [Sat, 18 Jun 2016 19:30:36 +0000 (12:30 -0700)]
nir/algebraic: Remove imprecise flog2 optimizations

While mathematically correct, these two optimizations result in an
expression with substantially lower precision than the original.  For any
positive finite floating-point value, log2(x) is well-defined and finite.
More precisely, it is in the range [-150, 150] so any sum of logarithms
log2(a) + log2(b) is also well-defined and finite as long as a and b are
both positive and finite.  However, if a and b are either very small or
very large, their product may get flushed to infinity or zero causing
log2(a * b) to be nowhere close to log2(a) + log2(b).

This imprecision was causing incorrect rendering in Talos Principal because
part of its HDR rendering process involves doing 8 texture operations,
clamping the result to [0, 65000], taking a dot-product with a constant,
and then taking the log2.  This is done 6 or 8 times and summed to produce
the final result which is written to a red texture.  In cases where you
have a region of the screen that is very dark, it can end up getting a
result value of -inf which is not what is intended.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96425
Cc: "11.1 11.2 12.0" <mesa-stable@lists.freedesktop.org>
8 years agoi965: Delete redundant extension enables
Ian Romanick [Mon, 16 May 2016 19:11:57 +0000 (12:11 -0700)]
i965: Delete redundant extension enables

A nearly identical block already exists in the gen >= 6 block above.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
8 years agomesa: Fix incorrect "see also" comments
Ian Romanick [Mon, 13 Jun 2016 20:48:47 +0000 (13:48 -0700)]
mesa: Fix incorrect "see also" comments

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
8 years agomesa: Silence unused parameter warning
Ian Romanick [Mon, 13 Jun 2016 20:45:00 +0000 (13:45 -0700)]
mesa: Silence unused parameter warning

main/pipelineobj.c: In function ‘delete_pipelineobj_cb’:
main/pipelineobj.c:110:30: warning: unused parameter ‘id’ [-Wunused-parameter]
 delete_pipelineobj_cb(GLuint id, void *data, void *userData)
                              ^

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
8 years agogallium: make image_view const
Rob Clark [Tue, 7 Jun 2016 18:32:16 +0000 (14:32 -0400)]
gallium: make image_view const

Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium: make constant_buffer const
Rob Clark [Tue, 7 Jun 2016 18:29:01 +0000 (14:29 -0400)]
gallium: make constant_buffer const

Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium: make shader_buffers const
Rob Clark [Tue, 7 Jun 2016 16:49:37 +0000 (12:49 -0400)]
gallium: make shader_buffers const

Be consistent with the rest of the "set_xyz" state interfaces.

Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: use trapezoid distribution for tess on Fiji and Polaris
Nicolai Hähnle [Fri, 17 Jun 2016 18:08:42 +0000 (20:08 +0200)]
radeonsi: use trapezoid distribution for tess on Fiji and Polaris

This yields a small performance improvement in Unigine Heaven.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agoradeonsi/sid: add Fiji+ tesselation distribution mode
Nicolai Hähnle [Tue, 17 May 2016 15:12:16 +0000 (10:12 -0500)]
radeonsi/sid: add Fiji+ tesselation distribution mode

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agoradeonsi: emit PA_SC_RASTER_CONFIG_1 only once
Nicolai Hähnle [Fri, 17 Jun 2016 09:01:30 +0000 (11:01 +0200)]
radeonsi: emit PA_SC_RASTER_CONFIG_1 only once

It is the same for all SEs.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agoradeonsi: fix calculation of valid RB mask per SE
Nicolai Hähnle [Fri, 17 Jun 2016 08:48:53 +0000 (10:48 +0200)]
radeonsi: fix calculation of valid RB mask per SE

The old calculation treated too many RBs as disabled.

Cc: 11.0 11.1 11.2 12.0 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agoradeonsi: raise SI_PM4_MAX_DW
Nicolai Hähnle [Fri, 17 Jun 2016 08:30:44 +0000 (10:30 +0200)]
radeonsi: raise SI_PM4_MAX_DW

The old limit, introduced in commit afa752d3f03ac6697581ff5d324e8ac0512ef513,
was exceeded by 4 SE configurations which hit si_write_harvested_raster_configs.

Cc: 11.1 11.2 12.0 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agogallivm: don't use integer min/max sse intrinsics with llvm >= 3.9
Roland Scheidegger [Sun, 19 Jun 2016 01:56:11 +0000 (03:56 +0200)]
gallivm: don't use integer min/max sse intrinsics with llvm >= 3.9

Apparently, these are deprecated. There's some AutoUpgrade feature which
is supposed to promote these to cmp/select, which apparently doesn't work
with jit code. It is possible it's not actually even meant to work (see
the bug filed against llvm which couldn't provide an answer neither)
but in any case this is meant to be only temporary unless the intrinsics
are really illegal. So, just use the fallback code (which should be cmp/select,
we're actually doing cmp/sext/trunc/select, but in any case llvm 3.9 manages
to optimize this back to pmin/pmax in the end).

This addresses https://llvm.org/bugs/show_bug.cgi?id=28176

CC: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Tested-by: Vinson Lee <vlee@freedesktop.org>
Tested-by: Aaron Watry <awatry@gmail.com>
8 years agonvc0: don't make use of push hint if there are no non-const user vbos
Ilia Mirkin [Sun, 19 Jun 2016 04:28:36 +0000 (00:28 -0400)]
nvc0: don't make use of push hint if there are no non-const user vbos

This makes the check match up what we do on nv50 as well - there's no
point in switching over the push path if everything's in managed
buffers. This can happen when a shader uses a vertex without an enabled
array - we end up passing it a constant attribute.

This also has the effect of "fixing" some flickering in Talos. I have no
idea why. I've stared at the push logic forwards, backwards, and
sideways. By always forcing the push path (which is slow), the
flickering also goes away, but other rendering is still wrong
(specifically draw 383068 as identified in the bug). However by not
switching over to the push path, draw 383068 is correct.

Note that other flickering remains in Talos, like the red/green
walls/floors. This takes care of the shadow flickering though.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90513
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
8 years agogk104/ir: fix tex use generation to be more careful about eliding uses
Ilia Mirkin [Sat, 18 Jun 2016 19:22:09 +0000 (15:22 -0400)]
gk104/ir: fix tex use generation to be more careful about eliding uses

If we have a loop, instructions before the tex might be added as tex
uses, and those may in fact dominate all other uses of the tex results.
This however doesn't mean that we don't need a texbar after the tex.
Only check if uses dominate each other they are dominated by the tex.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96565
Fixes: 7752bbc44 (gk104/ir: simplify and fool-proof texbar algorithm)
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "11.2 12.0" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
8 years agonv50: add support for GL_EXT_window_rectangles
Ilia Mirkin [Sun, 12 Jun 2016 20:05:31 +0000 (16:05 -0400)]
nv50: add support for GL_EXT_window_rectangles

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
8 years agonvc0: add support for GL_EXT_window_rectangles
Ilia Mirkin [Sun, 12 Jun 2016 04:46:18 +0000 (00:46 -0400)]
nvc0: add support for GL_EXT_window_rectangles

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
8 years agost/mesa: add support for GL_EXT_window_rectangles
Ilia Mirkin [Sat, 11 Jun 2016 20:52:17 +0000 (16:52 -0400)]
st/mesa: add support for GL_EXT_window_rectangles

Make sure to pass the requisite information in draws, blits, and clears
that work on the context's draw buffer.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium: add PIPE_CAP_MAX_WINDOW_RECTANGLES to all drivers
Ilia Mirkin [Sat, 11 Jun 2016 19:26:45 +0000 (15:26 -0400)]
gallium: add PIPE_CAP_MAX_WINDOW_RECTANGLES to all drivers

This says how many window rectangles are supported by the
implementation, although it may not exceed PIPE_MAX_WINDOW_RECTANGLES.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Brian Paul <brianp@vmware.com>
8 years agogallium: add API for setting window rectangles
Ilia Mirkin [Sat, 11 Jun 2016 15:35:01 +0000 (11:35 -0400)]
gallium: add API for setting window rectangles

Window rectangles apply to all framebuffer operations, either in
inclusive or exclusive mode. They may also be specified as part of a
blit operation.

In exclusive mode, any fragment inside any of the specified rectangles
will be discarded.

In inclusive mode, any fragment outside every rectangle will be
discarded.

The no-op state is to have 0 rectangles in exclusive mode.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agomesa: add GL_EXT_window_rectangles state storage/retrieval functionality
Ilia Mirkin [Fri, 10 Jun 2016 02:50:43 +0000 (22:50 -0400)]
mesa: add GL_EXT_window_rectangles state storage/retrieval functionality

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Brian Paul <brianp@vmware.com>
8 years agoglapi: add GL_EXT_window_rectangles entrypoints
Ilia Mirkin [Fri, 10 Jun 2016 02:10:10 +0000 (22:10 -0400)]
glapi: add GL_EXT_window_rectangles entrypoints

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Brian Paul <brianp@vmware.com>
8 years agonv50/ir: add missing strings for some recent sysvals
Samuel Pitoiset [Wed, 15 Jun 2016 14:14:39 +0000 (16:14 +0200)]
nv50/ir: add missing strings for some recent sysvals

This is pretty useful for debugging purposes and those should
not be omitted.

Fixes: 517a93b3 ("nvc0: add ARB_shader_draw_parameters support")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
8 years agoswr: Update screen->context pointer with multiple contexts.
Bruce Cherniak [Fri, 17 Jun 2016 17:14:54 +0000 (12:14 -0500)]
swr: Update screen->context pointer with multiple contexts.

A pipe pointer in the screen allows for access to current device context
 in flush_frontbuffer and resource_destroy.  This wasn't tracking current
context in multi-context situations.

v2: More caffeine.  Corrected compare, removed unnecessary set of
screen-pipe in create_context, and added a few comments.

8 years agoscons: put the generated git_sha1.h file in top-level src/ directory
Brian Paul [Wed, 15 Jun 2016 16:42:34 +0000 (10:42 -0600)]
scons: put the generated git_sha1.h file in top-level src/ directory

To match what's done in the automake build.

v2: Use git rev-parse to get a 10-character hash ID
    Fix Python imports

Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
8 years agoswr: switch from overriding -march to selecting features
Tim Rowley [Thu, 16 Jun 2016 19:56:39 +0000 (14:56 -0500)]
swr: switch from overriding -march to selecting features

Acked-by: Chuck Atkins <chuck.atkins@kitware.com>
Tested-by: Chuck Atkins <chuck.atkins@kitware.com>
8 years agomesa: remove remaining tabs in api_validate.c
Timothy Arceri [Fri, 17 Jun 2016 05:12:38 +0000 (15:12 +1000)]
mesa: remove remaining tabs in api_validate.c

Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
8 years agoi965/fs: indirect addressing with doubles is not supported in CHV/BSW/BXT
Samuel Iglesias Gonsálvez [Mon, 13 Jun 2016 06:29:53 +0000 (08:29 +0200)]
i965/fs: indirect addressing with doubles is not supported in CHV/BSW/BXT

From the Cherryview's PRM, Volume 7, 3D Media GPGPU Engine, Register Region
Restrictions, page 844:

  "When source or destination datatype is 64b or operation is integer DWord
   multiply, indirect addressing must not be used."

v2:
- Fix it for Broxton too.

v3:
- Simplify code by using subscript() and not creating a new num_components
variable (Kenneth).

Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95462
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
8 years agoi965/fs: Fix single-precision to double-precision conversions for CHV/BSW/BXT
Iago Toral Quiroga [Mon, 13 Jun 2016 07:13:23 +0000 (03:13 -0400)]
i965/fs: Fix single-precision to double-precision conversions for CHV/BSW/BXT

From the Cherryview PRM, Volume 7, 3D Media GPGPU Engine,
Register Region Restrictions:

   "When source or destination is 64b (...), regioning in Align1
    must follow these rules:

    1. Source and destination horizontal stride must be aligned to
       the same qword.
    (...)"

v2:
- Fix it for Broxton too.

v3:
- Remove inst->regs_written change as it is not necessary (Ken)

Cc: "12.0" <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95462
Tested-by: Mark Janes <mark.a.janes@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
8 years agodocs: Mention GL_ARB_ES3_1_compatibility in release notes.
Kenneth Graunke [Fri, 17 Jun 2016 00:10:08 +0000 (17:10 -0700)]
docs: Mention GL_ARB_ES3_1_compatibility in release notes.

Ilia reminded me that I forgot this.

8 years agoi965: Fix comment about CS scratch space encodings on Broadwell+.
Kenneth Graunke [Tue, 14 Jun 2016 08:35:20 +0000 (01:35 -0700)]
i965: Fix comment about CS scratch space encodings on Broadwell+.

I typo'd this.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
8 years agodocs: Update ARB_ES3_1_compatibility status for i965.
Kenneth Graunke [Thu, 16 Jun 2016 21:39:44 +0000 (14:39 -0700)]
docs: Update ARB_ES3_1_compatibility status for i965.

8 years agoi965: Drop perf_debug about rasterizer discard in SOL vs. clipper.
Kenneth Graunke [Wed, 11 May 2016 00:54:45 +0000 (17:54 -0700)]
i965: Drop perf_debug about rasterizer discard in SOL vs. clipper.

I recently experimented with performing rasterizer discard in the SOL
unit instead of the clipper, and as far as I can tell, it's basically
the same performance.  The clipper comes directly after SOL anyway,
and setting the clipper to REJECT_ALL should be pretty darn cheap.

Keep the perf_debug on Sandybridge, where the GS actually does work.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
8 years agoi965: Enable GL_ARB_ES3_1_compatibility on Gen8+ if CS are available.
Kenneth Graunke [Fri, 13 May 2016 01:08:49 +0000 (18:08 -0700)]
i965: Enable GL_ARB_ES3_1_compatibility on Gen8+ if CS are available.

There are almost no tests in any test suite, but what little I've found
seems to work.  Ilia believes everything is in place.

v2: Predicate the enable on ES 3.1 being available (Gen8+) and also
    ARB_compute_shader being available (requested by Ilia).

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>