Files
OpenColorIO/tests/gpu/GammaOp_test.cpp
caa20dd196 Adsk contrib - Add support for neon intrinsic integration (#1828)
* Merging the previous ARM Neon branch

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Testing each SIMD variants using a small code snippet and first pass on unifying the way related cmake variables

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Removing the usage of USE_SSE in favor of the new OCIO_USE_SSE2 as they serve the same purpose

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Using try_compile instead of check_cxx_source_compiles as it was given false positive.  Stubbing cpuinfo for Apple ARM plateform. Handling Apple M1 correctly and adding support for SSE2NEON.

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Comments clean up and refactor some comments and documentations

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Added something in the documentation for Rosetta and small change in cmake

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Fixing the build under Rosetta and fixing issue in the MacOS CI.

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Using try_compile for CheckSupportSSEUsingSSE2NEON to standardize the code and fixed CheckSupport compiler flags.

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* use emmintrin.h for only sse2 features

Signed-off-by: Mark Reid <mindmark@gmail.com>
Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Allow F16C to be completely turned off

Signed-off-by: Mark Reid <mindmark@gmail.com>
Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Seperate SIMD test code from code that adds unit tests.

This isolates the compulation units and
avoids executing illegal hardware instructions

Signed-off-by: Mark Reid <mindmark@gmail.com>
Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* remove uneeded checks

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* use software implementations of f16c intrinsics for SSE2

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Added preprocessor checks for ARM as it is needed for universal build on APPLE platform

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Adding missing checks for "not arm64"

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Ease the future maintainability a the new OCIO_USE_xyz be moving the logic into CPUInfoConfig.h.in as well as fixing issue on ARM when building with OCIO_USE_SSE2NEON=OFF.

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Fixing some spacing, documentations and making some cmake conditions clearer.

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Adding a build in ci_workflow for macos USE_OCIO_SSE2NEON=OFF

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* typo

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Changing back all the macOS (except one) builds to x86_64 only as it takes double the time to do the universal build.
OCIO no longuer build a universal binary by default

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Update the CMakeLists.txt logic to accomodate all scenario and fixing documentations.

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

* Update documentation and remove ocio_use_sse2neon from the CI matrix

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>

---------

Signed-off-by: Cédrik Fuoco <cedrik.fuoco@autodesk.com>
Signed-off-by: Mark Reid <mindmark@gmail.com>
Signed-off-by: Cédrik Fuoco <105517825+cedrik-fuoco-adsk@users.noreply.github.com>
Co-authored-by: Mark Reid <mindmark@gmail.com>
Co-authored-by: Doug Walker <doug.walker@autodesk.com>
2023-08-31 21:10:29 -04:00

239 lines
6.1 KiB
C++

// SPDX-License-Identifier: BSD-3-Clause
// Copyright Contributors to the OpenColorIO Project.
#include <OpenColorIO/OpenColorIO.h>
#include "GPUUnitTest.h"
namespace OCIO = OCIO_NAMESPACE;
const float g_epsilon = 1e-6f;
enum Version : unsigned
{
OCIO_VERSION_1 = 1,
OCIO_VERSION_2 = 2 // version 2 or higher
};
// Helper method to build unit tests
void AddExponent(OCIOGPUTest & test,
OCIO::TransformDirection direction,
const double(&gamma)[4],
OCIO::NegativeStyle style,
float epsilon,
Version ver)
{
OCIO::ExponentTransformRcPtr exp = OCIO::ExponentTransform::Create();
exp->setNegativeStyle(style);
exp->setDirection(direction);
exp->setValue(gamma);
OCIO_NAMESPACE::ConfigRcPtr config = OCIO_NAMESPACE::Config::Create();
config->setMajorVersion(ver);
test.setErrorThreshold(epsilon);
test.setProcessor(config, exp);
}
// Helper method to build unit tests
void AddExponentWithLinear(OCIOGPUTest & test,
OCIO::TransformDirection direction,
const double(&gamma)[4],
const double(&offset)[4],
OCIO::NegativeStyle style,
float epsilon)
{
OCIO::ExponentWithLinearTransformRcPtr exp = OCIO::ExponentWithLinearTransform::Create();
exp->setDirection(direction);
exp->setGamma(gamma);
exp->setOffset(offset);
exp->setNegativeStyle(style);
OCIO_NAMESPACE::ConfigRcPtr config = OCIO_NAMESPACE::Config::Create();
test.setErrorThreshold(epsilon);
test.setProcessor(config, exp);
}
OCIO_ADD_GPU_TEST(ExponentOp, legacy_shader_v1)
{
const double exp[4] = { 2.6, 1.0, 1.8, 1.1 };
AddExponent(test, OCIO::TRANSFORM_DIR_FORWARD, exp, OCIO::NEGATIVE_CLAMP, 1e-5f,
OCIO_VERSION_1);
test.setLegacyShader(true);
// TODO: Would like to be able to remove the setTestNaN(false) and
// setTestInfinity(false) from all of these tests.
test.setTestNaN(false);
}
OCIO_ADD_GPU_TEST(ExponentOp, forward_v1)
{
const double exp[4] = { 2.6, 1.0, 1.8, 1.1 };
AddExponent(test, OCIO::TRANSFORM_DIR_FORWARD, exp, OCIO::NEGATIVE_CLAMP, 1e-5f,
OCIO_VERSION_1);
test.setTestNaN(false);
}
OCIO_ADD_GPU_TEST(ExponentOp, forward)
{
const double exp[4] = { 2.6, 1.0, 1.8, 1.1 };
AddExponent(test, OCIO::TRANSFORM_DIR_FORWARD, exp, OCIO::NEGATIVE_CLAMP,
#if OCIO_USE_SSE2
5e-4f
#else
1e-5f
#endif
, OCIO_VERSION_2);
}
OCIO_ADD_GPU_TEST(ExponentOp, forward_mirror)
{
const double exp[4] = { 2.6, 1.0, 1.8, 1.1 };
AddExponent(test, OCIO::TRANSFORM_DIR_FORWARD, exp, OCIO::NEGATIVE_MIRROR,
#if OCIO_USE_SSE2
5e-4f // TODO: Only related to the ssePower optimization ?
#else
1e-5f
#endif
, OCIO_VERSION_2);
}
OCIO_ADD_GPU_TEST(ExponentOp, forward_pass_thru)
{
const double exp[4] = { 2.6, 1.0, 1.8, 1.1 };
AddExponent(test, OCIO::TRANSFORM_DIR_FORWARD, exp, OCIO::NEGATIVE_PASS_THRU,
#if OCIO_USE_SSE2
5e-4f // TODO: Only related to the ssePower optimization ?
#else
1e-5f
#endif
, OCIO_VERSION_2);
}
OCIO_ADD_GPU_TEST(ExponentOp, inverse_legacy_shader_v1)
{
const double exp[4] = { 2.6, 1.0, 1.8, 1.1 };
AddExponent(test, OCIO::TRANSFORM_DIR_INVERSE, exp, OCIO::NEGATIVE_CLAMP,
g_epsilon, OCIO_VERSION_1);
test.setLegacyShader(true);
test.setTestNaN(false);
}
OCIO_ADD_GPU_TEST(ExponentOp, inverse_v1)
{
const double exp[4] = { 2.6, 1.0, 1.8, 1.1 };
AddExponent(test, OCIO::TRANSFORM_DIR_INVERSE, exp, OCIO::NEGATIVE_CLAMP,
g_epsilon, OCIO_VERSION_1);
test.setTestNaN(false);
}
OCIO_ADD_GPU_TEST(ExponentOp, inverse)
{
const double exp[4] = { 2.6, 1.0, 1.8, 1.1 };
AddExponent(test, OCIO::TRANSFORM_DIR_INVERSE, exp, OCIO::NEGATIVE_CLAMP,
#if OCIO_USE_SSE2
5e-4f // TODO: Only related to the ssePower optimization ?
#else
g_epsilon
#endif
, OCIO_VERSION_2);
test.setTestInfinity(false);
}
OCIO_ADD_GPU_TEST(ExponentOp, inverse_mirror)
{
const double exp[4] = { 2.6, 1.0, 1.8, 1.1 };
AddExponent(test, OCIO::TRANSFORM_DIR_INVERSE, exp, OCIO::NEGATIVE_MIRROR,
#if OCIO_USE_SSE2
5e-4f // TODO: Only related to the ssePower optimization ?
#else
g_epsilon
#endif
, OCIO_VERSION_2);
test.setTestInfinity(false);
}
OCIO_ADD_GPU_TEST(ExponentOp, inverse_pass_thru)
{
const double exp[4] = { 2.6, 1.0, 1.8, 1.1 };
AddExponent(test, OCIO::TRANSFORM_DIR_INVERSE, exp, OCIO::NEGATIVE_PASS_THRU,
#if OCIO_USE_SSE2
5e-4f // TODO: Only related to the ssePower optimization ?
#else
g_epsilon
#endif
, OCIO_VERSION_2);
test.setTestInfinity(false);
}
static constexpr double gammaVals[4] = { 2.1, 1.0, 2.3, 1.5 };
static constexpr double offsetVals[4] = { .01, 0., .03, .05 };
OCIO_ADD_GPU_TEST(ExponentWithLinearOp, forward)
{
AddExponentWithLinear(test, OCIO::TRANSFORM_DIR_FORWARD, gammaVals, offsetVals,
OCIO::NEGATIVE_LINEAR,
#if OCIO_USE_SSE2
1e-4f // Note: Related to the ssePower optimization !
#else
5e-6f
#endif
);
test.setTestInfinity(false);
}
OCIO_ADD_GPU_TEST(ExponentWithLinearOp, mirror_forward)
{
AddExponentWithLinear(test, OCIO::TRANSFORM_DIR_FORWARD, gammaVals, offsetVals,
OCIO::NEGATIVE_MIRROR,
#if OCIO_USE_SSE2
1e-4f // Note: Related to the ssePower optimization !
#else
5e-6f
#endif
);
test.setTestInfinity(false);
}
OCIO_ADD_GPU_TEST(ExponentWithLinearOp, inverse)
{
AddExponentWithLinear(test, OCIO::TRANSFORM_DIR_INVERSE, gammaVals, offsetVals,
OCIO::NEGATIVE_LINEAR,
#if OCIO_USE_SSE2
5e-5f // Note: Related to the ssePower optimization !
#else
5e-7f
#endif
);
test.setTestInfinity(false);
}
OCIO_ADD_GPU_TEST(ExponentWithLinearOp, mirror_inverse)
{
AddExponentWithLinear(test, OCIO::TRANSFORM_DIR_INVERSE, gammaVals, offsetVals,
OCIO::NEGATIVE_MIRROR,
#if OCIO_USE_SSE2
5e-5f // Note: Related to the ssePower optimization !
#else
5e-7f
#endif
);
test.setTestInfinity(false);
}