mesa.git
8 years agoradeonsi: add GLSL lit tests
Marek Olšák [Sat, 30 Jul 2016 22:50:14 +0000 (00:50 +0200)]
radeonsi: add GLSL lit tests

They can only be run manually as described in HOW_TO_RUN.
It should help catch suboptimal code generation.

Some of the tests already fail.

v2: rename the tests to *.glsl,
    fix lit.cfg to find FileCheck

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com> (v1)
8 years agoradeonsi: add a standalone compiler amdgcn_glslc
Marek Olšák [Sat, 30 Jul 2016 22:48:11 +0000 (00:48 +0200)]
radeonsi: add a standalone compiler amdgcn_glslc

This will be used by GLSL lit tests.

For developers only. It shouldn't be distributable and it doesn't use
the Mesa build system.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: add environment variable SI_FORCE_FAMILY
Marek Olšák [Sat, 30 Jul 2016 22:46:09 +0000 (00:46 +0200)]
radeonsi: add environment variable SI_FORCE_FAMILY

This will be used by: amdgcn_glslc -mcpu=[family]

It can also be used for shader-db if you want stats for a different family.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agowinsys/radeon: implement cs_get_next_fence
Marek Olšák [Mon, 1 Aug 2016 22:44:55 +0000 (00:44 +0200)]
winsys/radeon: implement cs_get_next_fence

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agowinsys/amdgpu: implement cs_get_next_fence
Marek Olšák [Mon, 1 Aug 2016 22:44:55 +0000 (00:44 +0200)]
winsys/amdgpu: implement cs_get_next_fence

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium/radeon: add cs_get_next_fence winsys callback
Marek Olšák [Mon, 1 Aug 2016 22:43:02 +0000 (00:43 +0200)]
gallium/radeon: add cs_get_next_fence winsys callback

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium/radeon: count contexts
Marek Olšák [Fri, 15 Jul 2016 21:17:43 +0000 (23:17 +0200)]
gallium/radeon: count contexts

We don't wanna use unflushed fences when we have multiple contexts.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium/radeon: count gfx IB flushes
Marek Olšák [Fri, 15 Jul 2016 21:16:08 +0000 (23:16 +0200)]
gallium/radeon: count gfx IB flushes

This will be used as a counter for whether fence_finish needs to flush
the IB.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium/radeon: move radeon_winsys::cs_memory_below_limit to drivers
Marek Olšák [Fri, 29 Jul 2016 19:53:23 +0000 (21:53 +0200)]
gallium/radeon: move radeon_winsys::cs_memory_below_limit to drivers

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium/radeon: inline radeon_winsys::query_memory_usage
Marek Olšák [Fri, 29 Jul 2016 19:41:23 +0000 (21:41 +0200)]
gallium/radeon: inline radeon_winsys::query_memory_usage

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium/radeon/winsyses: expose per-IB used_vram and used_gart to drivers
Marek Olšák [Fri, 29 Jul 2016 19:39:00 +0000 (21:39 +0200)]
gallium/radeon/winsyses: expose per-IB used_vram and used_gart to drivers

The following patches will use this.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium/radeon/winsyses: print CS submission error number
Marek Olšák [Fri, 29 Jul 2016 16:31:33 +0000 (18:31 +0200)]
gallium/radeon/winsyses: print CS submission error number

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: flush if constant, shader, and streamout buffers use too much memory
Marek Olšák [Fri, 29 Jul 2016 16:18:36 +0000 (18:18 +0200)]
radeonsi: flush if constant, shader, and streamout buffers use too much memory

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: flush if sampler views and images use too much memory
Marek Olšák [Fri, 29 Jul 2016 15:48:28 +0000 (17:48 +0200)]
radeonsi: flush if sampler views and images use too much memory

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: deal with high vertex buffer memory usage correctly
Marek Olšák [Fri, 29 Jul 2016 15:28:43 +0000 (17:28 +0200)]
radeonsi: deal with high vertex buffer memory usage correctly

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: take compute shader and dispatch indirect memory usage into account
Marek Olšák [Fri, 29 Jul 2016 14:33:50 +0000 (16:33 +0200)]
radeonsi: take compute shader and dispatch indirect memory usage into account

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: take scratch buffer and draw indirect memory usage into account
Marek Olšák [Fri, 29 Jul 2016 14:33:50 +0000 (16:33 +0200)]
radeonsi: take scratch buffer and draw indirect memory usage into account

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: check IB memory usage of CP DMA operations
Marek Olšák [Fri, 29 Jul 2016 14:20:19 +0000 (16:20 +0200)]
radeonsi: check IB memory usage of CP DMA operations

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium/radeon: add r600_resource::vram_usage and gart_usage
Marek Olšák [Fri, 29 Jul 2016 13:48:18 +0000 (15:48 +0200)]
gallium/radeon: add r600_resource::vram_usage and gart_usage

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agomesa: Copy bitmask of VBOs in the VAO on gl{Push,Pop}Attrib.
Mathias Fröhlich [Mon, 1 Aug 2016 04:55:35 +0000 (06:55 +0200)]
mesa: Copy bitmask of VBOs in the VAO on gl{Push,Pop}Attrib.

On gl{Push,Pop}Attrib(GL_CLIENT_VERTEX_ARRAY_BIT) take
care that gl_vertex_array_object::VertexAttribBufferMask
matches the bound buffer object in the
gl_vertex_array_object::VertexBinding array.

Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Fredrik Höglund <fredrik@kde.org>
8 years agoanv/gen7_pipeline: Set PixelShaderKillPixel for discards
Nanley Chery [Thu, 4 Aug 2016 22:47:23 +0000 (15:47 -0700)]
anv/gen7_pipeline: Set PixelShaderKillPixel for discards

According to the IVB PRM Vol2 P1, this bit must be set if a pixel shader
contains a discard instruction.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97207
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
8 years agoutil/r11g11b10f: Whitespace cleanups
Jason Ekstrand [Wed, 3 Aug 2016 18:06:19 +0000 (11:06 -0700)]
util/r11g11b10f: Whitespace cleanups

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
8 years agoutil/format: Use explicitly sized types
Jason Ekstrand [Wed, 3 Aug 2016 16:58:13 +0000 (09:58 -0700)]
util/format: Use explicitly sized types

Both the rgb9e5 and r11g11b10 formats are defined based on how they are
packed into a 32-bit integer.  It makes sense that the functions that
manipulate them take an explicitly sized type.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
8 years agoutil/rgb9e5: Get rid of the float754 union
Jason Ekstrand [Wed, 3 Aug 2016 16:54:03 +0000 (09:54 -0700)]
util/rgb9e5: Get rid of the float754 union

There are a number of reasons for this refactor.  First, format_rgb9e5.h is
not something that a user would expect to define such a generic union.
Second, defining it requires checking for endianness which is ugly.  Third,
90% of what we were doing with the union was float <-> uint32_t bitcasts
and the remaining 10% can be done with a sinmple left-shift by 23.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
8 years agoutil/format_rgb9e5: Get rid of the rgb9e5 union
Jason Ekstrand [Wed, 3 Aug 2016 16:39:00 +0000 (09:39 -0700)]
util/format_rgb9e5: Get rid of the rgb9e5 union

The rgb9e5 format is a packed format defined in terms of slicing up a
single 32-bit value.  The bitfields are far more confusing than simple
shifts and require that we check the endianness.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
8 years agoutil: Move format_r11g11b10f.h to src/util
Jason Ekstrand [Wed, 3 Aug 2016 16:22:13 +0000 (09:22 -0700)]
util: Move format_r11g11b10f.h to src/util

It's used from both mesa main and gallium.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
8 years agoutil: Move format_rgb9e5.h to src/util
Jason Ekstrand [Wed, 3 Aug 2016 16:16:12 +0000 (09:16 -0700)]
util: Move format_rgb9e5.h to src/util

It's used from both mesa main and gallium.

Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
8 years agoglsl: fix indentation, comments and line lengths in ast_function.cpp
Andres Gomez [Tue, 2 Aug 2016 20:19:28 +0000 (23:19 +0300)]
glsl: fix indentation, comments and line lengths in ast_function.cpp

Acked-by: Timothy Arceri <timothy.arceri@collabora.com>
Signed-off-by: Andres Gomez <agomez@igalia.com>
8 years agoglsl: apply_implicit_conversion is static again
Andres Gomez [Tue, 2 Aug 2016 11:26:23 +0000 (14:26 +0300)]
glsl: apply_implicit_conversion is static again

Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Signed-off-by: Andres Gomez <agomez@igalia.com>
8 years agoglsl: struct constructors/initializers only allow implicit conversions
Andres Gomez [Sat, 30 Jul 2016 14:57:54 +0000 (17:57 +0300)]
glsl: struct constructors/initializers only allow implicit conversions

When an argument for a structure constructor or initializer doesn't
match the expected type, only Section 4.1.10 “Implicit Conversions”
are allowed to try to match that expected type.

From page 32 (page 38 of the PDF) of the GLSL 1.20 spec:

  " The arguments to the constructor will be used to set the structure's
    fields, in order, using one argument per field. Each argument must
    be the same type as the field it sets, or be a type that can be
    converted to the field's type according to Section 4.1.10 “Implicit
    Conversions.”"

From page 35 (page 41 of the PDF) of the GLSL 4.20 spec:

  " In all cases, the innermost initializer (i.e., not a list of
    initializers enclosed in curly braces) applied to an object must
    have the same type as the object being initialized or be a type that
    can be converted to the object's type according to section 4.1.10
    "Implicit Conversions". In the latter case, an implicit conversion
    will be done on the initializer before the assignment is done."

v2: Remove also the now redundant constant conversion, the
    constant_record_constructor helper and the replacement code
    (Timothy).

Fixes GL44-CTS.shading_language_420pack.initializer_list_negative

Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Signed-off-by: Andres Gomez <agomez@igalia.com>
8 years agoglsl: Refactor implicit conversion into its own helper
Andres Gomez [Tue, 2 Aug 2016 13:20:59 +0000 (16:20 +0300)]
glsl: Refactor implicit conversion into its own helper

v2: Refactor also the conversion to constant and replacement code
    (Timothy).

Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Signed-off-by: Andres Gomez <agomez@igalia.com>
8 years agoglsl/types: disallow implicit conversions before GLSL 1.20
Andres Gomez [Sat, 30 Jul 2016 22:46:59 +0000 (01:46 +0300)]
glsl/types: disallow implicit conversions before GLSL 1.20

Implicit conversions were added in the GLSL 1.20 spec version.

v2: Join the checks for GLSL 1.10 and ESSL (Timothy).

Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Signed-off-by: Andres Gomez <agomez@igalia.com>
8 years agoi965: Rework the unlit centroid workaround.
Kenneth Graunke [Tue, 2 Aug 2016 03:59:08 +0000 (20:59 -0700)]
i965: Rework the unlit centroid workaround.

Previously, for every input, we moved the dispatch mask to the flag
register, then emitted two predicated PLN instructions, one with
centroid barycentric coordinates (for normal pixels), and one with
pixel barycentric coordinates (for unlit helper pixels).

Instead, we can simply emit a set of predicated MOVs at the top of
the program which copy the pixel barycentric coordinates over the
centroid ones for unlit helper pixel channels.  Then, we can just
use normal PLNs.

On Sandybridge:

total instructions in shared programs: 7538470 -> 7534500 (-0.05%)
instructions in affected programs: 101268 -> 97298 (-3.92%)
helped: 705
HURT: 9 (all of which are SIMD16 programs)

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
8 years agoswr: [rasterizer core] static analysis fixes for conservative rast
Tim Rowley [Fri, 29 Jul 2016 17:47:49 +0000 (11:47 -0600)]
swr: [rasterizer core] static analysis fixes for conservative rast

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agoswr: [rasterizer core] implement InnerConservative input coverage
Tim Rowley [Thu, 28 Jul 2016 22:25:09 +0000 (16:25 -0600)]
swr: [rasterizer core] implement InnerConservative input coverage

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agoswr: [rasterizer core] remove CanEarlyZ function
Tim Rowley [Thu, 28 Jul 2016 19:16:46 +0000 (13:16 -0600)]
swr: [rasterizer core] remove CanEarlyZ function

Test is now in SetupPipeline.

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agoswr: [rasterizer core] use 32x32 macrotile for openswr
Tim Rowley [Tue, 26 Jul 2016 19:14:45 +0000 (13:14 -0600)]
swr: [rasterizer core] use 32x32 macrotile for openswr

Significant performance increase (up to 2x) on high geometry workloads.

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agoswr: [rasterizer fetch] add support for 24bit format fetch
Tim Rowley [Tue, 26 Jul 2016 18:32:14 +0000 (12:32 -0600)]
swr: [rasterizer fetch] add support for 24bit format fetch

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agoswr: [rasterizer fetch] additional fetch format support
Tim Rowley [Tue, 26 Jul 2016 18:10:36 +0000 (12:10 -0600)]
swr: [rasterizer fetch] additional fetch format support

Add support for 0 pitch in fetch.

Add support for USCALE/SSCALE for 32bit integer fetches.

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agoswr: [rasterizer jitter] fix potential jit exit crash
Tim Rowley [Mon, 25 Jul 2016 23:05:55 +0000 (17:05 -0600)]
swr: [rasterizer jitter] fix potential jit exit crash

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agoswr: [rasterizer core] update sync handling
Tim Rowley [Mon, 25 Jul 2016 22:24:58 +0000 (16:24 -0600)]
swr: [rasterizer core] update sync handling

Sync now uses a callback to ensure that it's called by the last
thread moving past a DC.  This will help with the new counter
handling.

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agoswr: [rasterizer core] rename variable
Tim Rowley [Mon, 25 Jul 2016 22:21:47 +0000 (16:21 -0600)]
swr: [rasterizer core] rename variable

Avoid nested declarations of the same name within a single function.

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agoswr: [rasterizer jitter] adjust extern "C" block scope
Tim Rowley [Thu, 21 Jul 2016 22:54:25 +0000 (16:54 -0600)]
swr: [rasterizer jitter] adjust extern "C" block scope

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agoswr: [rasterizer core] conservative rast degenerate handling
Tim Rowley [Wed, 20 Jul 2016 22:05:25 +0000 (16:05 -0600)]
swr: [rasterizer core] conservative rast degenerate handling

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agoswr: [rasterizer core] allow hexadecimal for integer knobs
Tim Rowley [Tue, 19 Jul 2016 18:51:46 +0000 (12:51 -0600)]
swr: [rasterizer core] allow hexadecimal for integer knobs

Signed-off-by: Tim Rowley <timothy.o.rowley@intel.com>
8 years agomesa: Dynamically allocate the matrix stack.
Eric Anholt [Tue, 2 Aug 2016 19:48:47 +0000 (12:48 -0700)]
mesa: Dynamically allocate the matrix stack.

By allocating and initializing the matrices at context creation, the OS
couldn't even overcommit the pages.  This saves about 63k (out of 946k) of
maximum memory size according to massif on simulated vc4
glsl-algebraic-add-add-1.  It also means we could potentially relax the
maximum stack sizes, but that should be a separate commit.

v2: Drop redundant Top update, explain why the stack is small at init
    time.

Reviewed-by: Brian Paul <brianp@vmware.com>
8 years agostate_tracker: Initialize the draw context only when needed.
Eric Anholt [Tue, 2 Aug 2016 19:28:27 +0000 (12:28 -0700)]
state_tracker: Initialize the draw context only when needed.

It's only used for rarely-used deprecated GL features
(feedback/rasterpos), so we can skip the memory allocation and
initialization for it most of the time.

Saves about 659k (out of 1605k) of maximum memory size according to massif
on simulated vc4 glsl-algebraic-add-add-1

Reviewed-by: Brian Paul <brianp@vmware.com>
8 years agovc4: Move scalarizing and some lowering to link time.
Eric Anholt [Wed, 3 Aug 2016 21:01:03 +0000 (14:01 -0700)]
vc4: Move scalarizing and some lowering to link time.

This works out to be a wash in terms of memory usage: We use more memory
to store the separate ALU instructions, but we optimize out a lot of code
as well.  The main result, though, is that we do more of our work at link
time rather than draw time.

8 years agovc4: Avoid VS shader recompiles by keeping a set of FS inputs seen so far.
Eric Anholt [Wed, 3 Aug 2016 18:55:55 +0000 (11:55 -0700)]
vc4: Avoid VS shader recompiles by keeping a set of FS inputs seen so far.

We don't want to bake the whole array into the FS key, because of the
hashing overhead.  But we can keep a set of the arrays seen, and use a
pointer to the copy in as the array's proxy.

Between this and the previous patch, gl-1.0-blend-func now passes on
hardware, where previously it was filling the 256MB CMA area with shaders
and OOMing.

Drops 712 shaders from shader-db.

8 years agovc4: Don't recompile the CS when the FS changes.
Eric Anholt [Wed, 3 Aug 2016 18:57:53 +0000 (11:57 -0700)]
vc4: Don't recompile the CS when the FS changes.

The compiled_fs_id is a proxy for the vc4->prog.fs->input_slots[], but
only the VS dereferences it.

Drops 754 shaders from shader-db.

8 years agovc4: Move FS inputs setup out to a helper function.
Eric Anholt [Wed, 3 Aug 2016 18:39:59 +0000 (11:39 -0700)]
vc4: Move FS inputs setup out to a helper function.

It's a pretty big block, and I was about to make it bigger.

8 years agonir: Make nir_opt_remove_phis see through moves.
Kenneth Graunke [Fri, 29 Jul 2016 08:29:12 +0000 (01:29 -0700)]
nir: Make nir_opt_remove_phis see through moves.

I found a shader in Tales of Maj'Eyal that contains:

        if ssa_21 {
                block block_1:
                /* preds: block_0 */
                ...instructions that prevent the select peephole...
                vec1 32 ssa_23 = imov ssa_4
                vec1 32 ssa_24 = imov ssa_4.y
                vec1 32 ssa_25 = imov ssa_4.z
                /* succs: block_3 */
        } else {
                block block_2:
                /* preds: block_0 */
                vec1 32 ssa_26 = imov ssa_4
                vec1 32 ssa_27 = imov ssa_4.y
                vec1 32 ssa_28 = imov ssa_4.z
                /* succs: block_3 */
        }
        block block_3:
        /* preds: block_1 block_2 */
        vec1 32 ssa_29 = phi block_1: ssa_23, block_2: ssa_26
        vec1 32 ssa_30 = phi block_1: ssa_24, block_2: ssa_27
        vec1 32 ssa_31 = phi block_1: ssa_25, block_2: ssa_28

Here, copy propagation will bail because phis cannot perform swizzles,
and CSE won't do anything because there is no dominance relationship
between the imovs.  By making nir_opt_remove_phis handle identical moves,
we can eliminate the phis and rewrite everything to use ssa_4 directly,
so all the moves become dead and get eliminated.

I don't think we need to check "exact" - just the alu sources.
Presumably phi sources should match in their exactness.

On Broadwell:

total instructions in shared programs: 11639872 -> 11638535 (-0.01%)
instructions in affected programs: 134222 -> 132885 (-1.00%)
helped: 338
HURT: 0

v2: Fix return value to be NULL, not false (caught by Iago).

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
8 years agonir: Make nir_alu_srcs_equal non-static.
Kenneth Graunke [Fri, 29 Jul 2016 08:29:11 +0000 (01:29 -0700)]
nir: Make nir_alu_srcs_equal non-static.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
8 years agonir: Turn imov/fmov of undef into undef.
Kenneth Graunke [Fri, 29 Jul 2016 08:29:10 +0000 (01:29 -0700)]
nir: Turn imov/fmov of undef into undef.

On Broadwell:

total instructions in shared programs: 11640214 -> 11639872 (-0.00%)
instructions in affected programs: 17744 -> 17402 (-1.93%)
helped: 78
HURT: 0

total spills in shared programs: 2924 -> 2922 (-0.07%)
spills in affected programs: 104 -> 102 (-1.92%)
helped: 1
HURT: 0

total fills in shared programs: 4394 -> 4389 (-0.11%)
fills in affected programs: 237 -> 232 (-2.11%)
helped: 1
HURT: 0

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
8 years agoi965: Use a separate register for every access to an SSA undef.
Kenneth Graunke [Fri, 29 Jul 2016 08:29:09 +0000 (01:29 -0700)]
i965: Use a separate register for every access to an SSA undef.

Previously, we allocated a new VGRF for every undefined definition.
Instead, this patch makes us allocate a new VGRF for every use of an
undefined definition.  This makes sure that undefined values are
fully independent of one another, and have live ranges limited to
their single use.  This allows register coalescing to combine the
source and destination of MOVs from undefined sources, eliminating
the MOV altogether.

On Broadwell:

total instructions in shared programs: 11641187 -> 11640214 (-0.01%)
instructions in affected programs: 70199 -> 69226 (-1.39%)
helped: 213
HURT: 1

v2: Add a comment (based on Iago's suggested one).

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
8 years agovl/dri3: Destroy Present event context when destroying drawable v2
Michel Dänzer [Thu, 28 Jul 2016 08:52:11 +0000 (17:52 +0900)]
vl/dri3: Destroy Present event context when destroying drawable v2

Without this, the X server may accumulate stale Present event contexts
if a client performs several video decoding sessions using the same
window.

v2: Based on Chris Wilson's review:
* Use xcb_discard_reply() instead of free(xcb_request_check())

Reviewed-and-Tested-by: Leo Liu <leo.liu@amd.com>
8 years agoloader/dri3: Destroy Present event context when destroying drawable v2
Michel Dänzer [Thu, 28 Jul 2016 08:44:49 +0000 (17:44 +0900)]
loader/dri3: Destroy Present event context when destroying drawable v2

Without this, the X server may accumulate stale Present event contexts
if a client ends up creating and destroying DRI drawables for the same
window.

v2: Based on Chris Wilson's review:
* Use xcb_present_select_input_checked so that protocol errors
  generated by old X servers can be handled gracefully
* Use xcb_discard_reply() instead of free(xcb_request_check())

8 years agogbm: Correct bo_import documentation (trivial)
Ben Widawsky [Tue, 2 Aug 2016 00:18:48 +0000 (17:18 -0700)]
gbm: Correct bo_import documentation (trivial)

Missed here:
commit a43d286ef7ff65087b1f051d071b829ca7b02073
Author: Kristian Høgsberg <krh@bitplanet.net>
Date:   Fri Mar 28 10:17:11 2014 -0700

    gbm: Add import from fd

Cc: Kristian Høgsberg <krh@bitplanet.net>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
8 years agovc4: Avoid generating a custom shader per level in glGenerateMipmaps().
Eric Anholt [Tue, 2 Aug 2016 23:01:34 +0000 (16:01 -0700)]
vc4: Avoid generating a custom shader per level in glGenerateMipmaps().

We were baking in the LOD of the source level to each shader.  Instead,
pass it in as a uniform -- this requires storing it to a temp register,
but that's better than compiling a ton of separate shaders:

total instructions in shared programs: 115032 -> 115036 (0.00%)
instructions in affected programs:     96 -> 100 (4.17%)
LOST:                                  572

8 years agovc4: Tell valgrind about BO allocations from mmap time to destroy.
Eric Anholt [Tue, 2 Aug 2016 22:01:59 +0000 (15:01 -0700)]
vc4: Tell valgrind about BO allocations from mmap time to destroy.

This helps in debugging memory pressure.  It would be nice if we could
tell valgrind about it all the way from allocation time to destroy, but we
need a pointer to hand to VALGRIND_MALLOCLIKE_BLOCK.

8 years agoloader: fix memory leak in loader_dri3_open
Jan Ziak [Sat, 30 Jul 2016 00:31:10 +0000 (02:31 +0200)]
loader: fix memory leak in loader_dri3_open

Found via "valgrind --leak-check=full glxgears".

Signed-off-by: Jan Ziak (http://atom-symbol.net) <0xe2.0x9a.0x9b@gmail.com>
Acked-by: Boyan Ding <boyan.j.ding@gmail.com>
Cc: "12.0 11.2" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
8 years agovc4: Fix a leak of the src[] array of VPM reads in optimization.
Eric Anholt [Tue, 2 Aug 2016 18:02:21 +0000 (11:02 -0700)]
vc4: Fix a leak of the src[] array of VPM reads in optimization.

Cc: "12.0" <mesa-stable@lists.freedesktop.org>
8 years agovc4: Fix leak of the bo_handles table.
Eric Anholt [Tue, 2 Aug 2016 17:57:01 +0000 (10:57 -0700)]
vc4: Fix leak of the bo_handles table.

8 years agovc4: Fix handling of UBO range offsets.
Eric Anholt [Wed, 27 Jul 2016 21:58:43 +0000 (14:58 -0700)]
vc4: Fix handling of UBO range offsets.

The ranges are in units of bytes, not dwords.  This wasn't caught by
piglit tests because ttn tends to make one big uniform file, so we only
had one UBO range with a src and dst offset of 0.

8 years agonir: Allow opt_peephole_select to work on empty blocks.
Eric Anholt [Thu, 28 Jul 2016 00:26:59 +0000 (17:26 -0700)]
nir: Allow opt_peephole_select to work on empty blocks.

nir_opt_peephole_select has the job of removing IF statements with no side
effects.  However, if the IF statement's successor didn't have any
instructions in it, we were skipping it, which occurred in mupen64 on vc4
with glsl_to_nir enabled:

instructions in affected programs:     6134 -> 4120 (-32.83%)
total uniforms in shared programs: 38268 -> 38219 (-0.13%)

No changes on Haswell shader-db.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
8 years agovc4: Dump NIR at shader state creation time as well.
Eric Anholt [Wed, 27 Jul 2016 00:21:46 +0000 (17:21 -0700)]
vc4: Dump NIR at shader state creation time as well.

I keep wanting to see this version of the NIR.

8 years agor600g: use last_gfx_fence like radeonsi
Marek Olšák [Thu, 14 Jul 2016 23:06:58 +0000 (01:06 +0200)]
r600g: use last_gfx_fence like radeonsi

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agogallium/radeon: move last_gfx_fence from radeonsi to common code
Marek Olšák [Thu, 14 Jul 2016 22:39:38 +0000 (00:39 +0200)]
gallium/radeon: move last_gfx_fence from radeonsi to common code

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: skip unnecessary si_update_shaders calls
Marek Olšák [Tue, 2 Aug 2016 09:51:21 +0000 (11:51 +0200)]
radeonsi: skip unnecessary si_update_shaders calls

Small decrease in draw call overhead.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: print the command line to VM fault reports (v2)
Marek Olšák [Sat, 30 Jul 2016 12:57:20 +0000 (14:57 +0200)]
radeonsi: print the command line to VM fault reports (v2)

v2: rebase on top of Brian's commit

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoddebug: print the command line to all logs (v2)
Marek Olšák [Sat, 30 Jul 2016 12:56:28 +0000 (14:56 +0200)]
ddebug: print the command line to all logs (v2)

for piglit with the pipelined hang detection mode

v2: rebase on top of Brian's commit

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoddebug: don't use fmemopen on non-Linux OS
Marek Olšák [Sat, 30 Jul 2016 21:54:06 +0000 (23:54 +0200)]
ddebug: don't use fmemopen on non-Linux OS

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97140

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: don't set the last parameter component of llvm.AMDGPU.cube
Marek Olšák [Sat, 30 Jul 2016 14:32:32 +0000 (16:32 +0200)]
radeonsi: don't set the last parameter component of llvm.AMDGPU.cube

LLVM doesn't use it.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: use llvm.amdgcn.cube* if available
Marek Olšák [Sat, 30 Jul 2016 14:28:20 +0000 (16:28 +0200)]
radeonsi: use llvm.amdgcn.cube* if available

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: use llvm.amdgcn.rsq.f64 if available
Marek Olšák [Sat, 30 Jul 2016 14:23:48 +0000 (16:23 +0200)]
radeonsi: use llvm.amdgcn.rsq.f64 if available

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoradeonsi: use v_mad_f32 for fma
Marek Olšák [Tue, 26 Jul 2016 20:34:03 +0000 (22:34 +0200)]
radeonsi: use v_mad_f32 for fma

v_fma_f32 runs at FP64 rate (= slow). Alien Isolation and F1 2015 seem
to use fma for all d3d multiply-add instructions, which is silly.

This tries to restore performance for those games.

The main difference between v_mad_f32 and v_fma_f32 is that v_mad doesn't
support denormals, which we don't enable anyway, because they are slow too.

Also, there is code size reduction:
  Totals from affected shaders:
  VGPRS: 109796 -> 109808 (0.01 %)
  Spilled SGPRs: 29995 -> 30022 (0.09 %)
  Spilled VGPRs: 12 -> 13 (8.33 %) <-- it's just one shader going from 12 to 13
  Code Size: 6667596 -> 6476356 (-2.87 %) bytes
  Max Waves: 26931 -> 26899 (-0.12 %)

I've not actually tested real performance.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
8 years agoi965: use mt->offset in intel_miptree_map_movntdqa()
Haixia Shi [Wed, 3 Aug 2016 00:03:32 +0000 (17:03 -0700)]
i965: use mt->offset in intel_miptree_map_movntdqa()

We need to include mt->offset in the calculation of src pointer because its
value may be non-zero, for example in a cubemap texture.

Signed-off-by: Haixia Shi <hshi@chromium.org>
Cc: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Chad Versace <chad@kiwitree.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Change-Id: I461ad5b204626d5a1c45611fc6b63735dcf29f63

8 years agonir: fix validation message
Timothy Arceri [Tue, 2 Aug 2016 07:46:19 +0000 (17:46 +1000)]
nir: fix validation message

Looks like a copy and paste error from f752effa087

Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
8 years ago.mailmap: Update my address
Chad Versace [Tue, 2 Aug 2016 20:29:53 +0000 (13:29 -0700)]
.mailmap: Update my address

I left Intel, so make my personal address the canonical address.

8 years agoswr: build swr with -fno-strict-aliasing
Tim Rowley [Tue, 2 Aug 2016 18:18:08 +0000 (13:18 -0500)]
swr: build swr with -fno-strict-aliasing

swr rasterizer contains numerous data transfers between vectors
and ordinary C types.  Fixing for strict aliasing will take time.

Reviewed-by: Matt Turner <mattst88@gmail.com>
8 years agoast: Updated AST_NUM_OPERATORS for coherence with ast_operators
Andres Gomez [Sun, 31 Jul 2016 16:07:34 +0000 (19:07 +0300)]
ast: Updated AST_NUM_OPERATORS for coherence with ast_operators

AST_NUM_OPERATORS stores the dimension of the ast_operators
enumeration but was not updated after its last modification.

This doesn't add any real modification for any code paths but it makes
sense for coherence.

v2 (Eric Engestrom): Just place the define at the end of the
                     enumeration, not below.

Signed-off-by: Andres Gomez <agomez@igalia.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
8 years agoi965: Disable the unlit centroid workaround on Gen7.
Matt Turner [Mon, 1 Aug 2016 19:18:45 +0000 (12:18 -0700)]
i965: Disable the unlit centroid workaround on Gen7.

Once upon a time (commit 8313f44409) Paul added code for the unlit
centroid workaround (WaCopyUnlitCentroidBarys). His commit message
claims it fixed the EXT_framebuffer_multisample/interpolation {2,4}
{centroid-deriv,centroid-deriv-disabled} piglit tests but does not say
on which platform, though he cites the IVB PRM.

"3DSTATE_WM [DevIVB, DevHSW]" says

   "[DevIVB]: Workaround: When Centroid Barycentric mode is required, HW
    may produce incorrect interpolation results when a 2X2 pixels have
    unlit pixels."

I later disabled it for Haswell (commit f6db414f3c) with no known ill
effects.

The Sandybridge page does not have this text, but the workarounds
database (see WaCopyUnlitCentroidBarys) says the issues applies *only*
to Sandybridge, and in fact in commit 1a2de7dce8fc I note that disabling
the workaround on Sandybridge causes the tests Paul originally mentioned
to fail.

So this is, and always has been, a huge confusing mess.

Disabling the workaround indeed causes the tests Paul originally
mentioned to fail on Sandybridge but not on Ivybridge/Baytrail.

On Ivybridge:

   total instructions in shared programs: 6914901 -> 6909599 (-0.08%)
   instructions in affected programs: 106766 -> 101464 (-4.97%)
   helped: 884

   total cycles in shared programs: 70874764 -> 70813774 (-0.09%)
   cycles in affected programs: 794144 -> 733154 (-7.68%)
   helped: 688
   HURT: 186

   LOST:   1
   GAINED: 6

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
8 years agogallium/util: fix align64
Marek Olšák [Fri, 29 Jul 2016 13:00:12 +0000 (15:00 +0200)]
gallium/util: fix align64

it cut off the upper 32 bits

Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
8 years agomesa: Drop -fno-strict-aliasing.
Matt Turner [Thu, 26 May 2016 23:34:02 +0000 (16:34 -0700)]
mesa: Drop -fno-strict-aliasing.

Improves performance of OglBatch7 by 4.06851% +/- 1.17925% (n=169) on
Haswell, and cuts ~18k of .text:

   text     data      bss      dec      hex  filename
5824627   287816    29384  6141827   5db783  before/i965_dri.so
5806354   287816    29384  6123554   5d7022  after/i965_dri.so

Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
8 years agoi915: Avoid aliasing violation.
Matt Turner [Thu, 28 Jul 2016 19:44:41 +0000 (12:44 -0700)]
i915: Avoid aliasing violation.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
8 years agodraw: Avoid aliasing violations.
Matt Turner [Thu, 28 Jul 2016 21:41:17 +0000 (14:41 -0700)]
draw: Avoid aliasing violations.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agor600g: Avoid aliasing violations.
Matt Turner [Thu, 28 Jul 2016 19:38:21 +0000 (12:38 -0700)]
r600g: Avoid aliasing violations.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agor300g: Avoid aliasing violation.
Matt Turner [Thu, 28 Jul 2016 20:57:06 +0000 (13:57 -0700)]
r300g: Avoid aliasing violation.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agogallium/auxiliary: Add u_bitcast.h header.
Matt Turner [Thu, 28 Jul 2016 21:34:14 +0000 (14:34 -0700)]
gallium/auxiliary: Add u_bitcast.h header.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agoglsl_to_tgsi: Avoid aliasing violations.
Matt Turner [Thu, 28 Jul 2016 19:43:19 +0000 (12:43 -0700)]
glsl_to_tgsi: Avoid aliasing violations.

Reviewed-by: Dave Airlie <airlied@redhat.com>
8 years agost/mesa: silence missing braces warning in st_program.c
Brian Paul [Mon, 1 Aug 2016 15:56:21 +0000 (09:56 -0600)]
st/mesa: silence missing braces warning in st_program.c

Silence a gcc warning:
state_tracker/st_program.c: In function 'st_create_fp_variant':
state_tracker/st_program.c:957:10: warning: missing braces around initializer [-Wmissing-braces]
          nir_lower_drawpixels_options options = {0};
          ^
state_tracker/st_program.c:957:10: warning: (near initialization for 'options.texcoord_state_tokens') [-Wmissing-braces]

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agoauxiliary/os: add new os_get_command_line() function
Brian Paul [Wed, 29 Jun 2016 22:30:54 +0000 (16:30 -0600)]
auxiliary/os: add new os_get_command_line() function

This can be used by the driver to get the command line which started
the process.  Will be used by the VMware driver for extra logging.

For now, this is only implemented for Linux via /proc/self/cmdline
and Windows via GetCommandLine().

Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agosvga: avoid redundant SetVertexBuffer/SetIndexBuffer commands at rebind
Charmaine Lee [Fri, 22 Jul 2016 22:05:02 +0000 (15:05 -0700)]
svga: avoid redundant SetVertexBuffer/SetIndexBuffer commands at rebind

This patch eliminates the redundant SetVertexBuffers and
SetIndexBuffer commands that are emitted for rebind purpose.
With this patch, the set commands will be skipped, but we will still
reference the associated resources to allow the kernel to
bring in the resources.

Tested with Lightsmark2008, Valley, MTT glretrace, piglit, conform.

Reviewed-by: Brian Paul <brianp@vmware.com>
8 years agou_vbuf: fix potentially bogus assert
Rob Clark [Sat, 30 Jul 2016 15:22:52 +0000 (11:22 -0400)]
u_vbuf: fix potentially bogus assert

There are cases where we hit u_vbuf path due to alignment or pitch-
alignment restrictions, but for an output-format that u_vbuf does not
support translating (yet the driver does support natively).  In which
case we hit the memcpy() path and don't care that u_vbuf doesn't
understand it.

Fixes crash with debug build of mesa in:
dEQP-GLES3.functional.vertex_arrays.single_attribute.strides.fixed.user_ptr_stride17_components2_quads1

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95000
Signed-off-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
8 years agogbm: Removed unused function.
Ben Widawsky [Fri, 29 Jul 2016 20:42:04 +0000 (13:42 -0700)]
gbm: Removed unused function.

AFAICT, it's never been used.

It was briefly nudged in the right direction here:
commit 10e5ffd4961055ebba5be4d85a93cc66cdd5a635
Author: Emil Velikov <emil.l.velikov@gmail.com>
Date:   Sat Jan 25 17:19:10 2014 +0000

    gbm: do not export _gbm_mesa_get_device

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
8 years agoi965: fix comparison warning
Timothy Arceri [Mon, 1 Aug 2016 00:35:06 +0000 (10:35 +1000)]
i965: fix comparison warning

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
8 years agovc4: Zero-initialize the hardware sampler view structure.
Eric Anholt [Tue, 26 Jul 2016 22:32:11 +0000 (15:32 -0700)]
vc4: Zero-initialize the hardware sampler view structure.

Fixes failure to initialize the force_first_level flag, causing
failures in piglit levelclamp.

8 years agomesa: Remove set but not used gl_client_array::Stride.
Mathias Fröhlich [Fri, 17 Jun 2016 06:09:05 +0000 (08:09 +0200)]
mesa: Remove set but not used gl_client_array::Stride.

The field is only read for printing today and
there it was probably a leftover.

Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Reviewed-by: Eric Anholt <eric@anholt.net>
8 years agomesa: Remove set but not used gl_client_array::Enabled.
Mathias Fröhlich [Fri, 17 Jun 2016 06:09:05 +0000 (08:09 +0200)]
mesa: Remove set but not used gl_client_array::Enabled.

The way it is used today does not care about the
Enabled flag anymore.

Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Reviewed-by: Eric Anholt <eric@anholt.net>
8 years agovbo: Use the VAO array enabled flags in vbo_exec_array.
Mathias Fröhlich [Fri, 17 Jun 2016 06:09:05 +0000 (08:09 +0200)]
vbo: Use the VAO array enabled flags in vbo_exec_array.

Instead of gl_client_array::Enabled inside a VAO,
directly use the gl_vertex_attrib_array::Enabled value
which is the origin of the above.

Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Reviewed-by: Eric Anholt <eric@anholt.net>