5c2696c378
This commit reworks the unit test for 8x8 forward/inverse transformation. It adds extreme input value test to detect overflow issues in the intermediate steps. It temporarily disables unit test for the SSSE3 version, which showed overflow failure in the new test conditions. Change-Id: I7caf10bba4b6db031add65d8c0eb99426b38aa42
282 lines
8.9 KiB
C++
282 lines
8.9 KiB
C++
/*
|
|
* Copyright (c) 2012 The WebM project authors. All Rights Reserved.
|
|
*
|
|
* Use of this source code is governed by a BSD-style license
|
|
* that can be found in the LICENSE file in the root of the source
|
|
* tree. An additional intellectual property rights grant can be found
|
|
* in the file PATENTS. All contributing project authors may
|
|
* be found in the AUTHORS file in the root of the source tree.
|
|
*/
|
|
|
|
#include <math.h>
|
|
#include <stdlib.h>
|
|
#include <string.h>
|
|
|
|
#include "third_party/googletest/src/include/gtest/gtest.h"
|
|
#include "test/acm_random.h"
|
|
#include "test/clear_system_state.h"
|
|
#include "test/register_state_check.h"
|
|
#include "test/util.h"
|
|
|
|
#include "./vpx_config.h"
|
|
#include "./vp9_rtcd.h"
|
|
#include "vp9/common/vp9_entropy.h"
|
|
#include "vpx/vpx_integer.h"
|
|
|
|
using libvpx_test::ACMRandom;
|
|
|
|
namespace {
|
|
#ifdef _MSC_VER
|
|
static int round(double x) {
|
|
if (x < 0)
|
|
return static_cast<int>(ceil(x - 0.5));
|
|
else
|
|
return static_cast<int>(floor(x + 0.5));
|
|
}
|
|
#endif
|
|
|
|
const int kNumCoeffs = 1024;
|
|
const double kPi = 3.141592653589793238462643383279502884;
|
|
void reference_32x32_dct_1d(const double in[32], double out[32], int stride) {
|
|
const double kInvSqrt2 = 0.707106781186547524400844362104;
|
|
for (int k = 0; k < 32; k++) {
|
|
out[k] = 0.0;
|
|
for (int n = 0; n < 32; n++)
|
|
out[k] += in[n] * cos(kPi * (2 * n + 1) * k / 64.0);
|
|
if (k == 0)
|
|
out[k] = out[k] * kInvSqrt2;
|
|
}
|
|
}
|
|
|
|
void reference_32x32_dct_2d(const int16_t input[kNumCoeffs],
|
|
double output[kNumCoeffs]) {
|
|
// First transform columns
|
|
for (int i = 0; i < 32; ++i) {
|
|
double temp_in[32], temp_out[32];
|
|
for (int j = 0; j < 32; ++j)
|
|
temp_in[j] = input[j*32 + i];
|
|
reference_32x32_dct_1d(temp_in, temp_out, 1);
|
|
for (int j = 0; j < 32; ++j)
|
|
output[j * 32 + i] = temp_out[j];
|
|
}
|
|
// Then transform rows
|
|
for (int i = 0; i < 32; ++i) {
|
|
double temp_in[32], temp_out[32];
|
|
for (int j = 0; j < 32; ++j)
|
|
temp_in[j] = output[j + i*32];
|
|
reference_32x32_dct_1d(temp_in, temp_out, 1);
|
|
// Scale by some magic number
|
|
for (int j = 0; j < 32; ++j)
|
|
output[j + i * 32] = temp_out[j] / 4;
|
|
}
|
|
}
|
|
|
|
typedef void (*fwd_txfm_t)(const int16_t *in, int16_t *out, int stride);
|
|
typedef void (*inv_txfm_t)(const int16_t *in, uint8_t *out, int stride);
|
|
|
|
typedef std::tr1::tuple<fwd_txfm_t, inv_txfm_t, int> trans_32x32_param_t;
|
|
|
|
class Trans32x32Test : public ::testing::TestWithParam<trans_32x32_param_t> {
|
|
public:
|
|
virtual ~Trans32x32Test() {}
|
|
virtual void SetUp() {
|
|
fwd_txfm_ = GET_PARAM(0);
|
|
inv_txfm_ = GET_PARAM(1);
|
|
version_ = GET_PARAM(2); // 0: high precision forward transform
|
|
// 1: low precision version for rd loop
|
|
}
|
|
|
|
virtual void TearDown() { libvpx_test::ClearSystemState(); }
|
|
|
|
protected:
|
|
int version_;
|
|
fwd_txfm_t fwd_txfm_;
|
|
inv_txfm_t inv_txfm_;
|
|
};
|
|
|
|
TEST_P(Trans32x32Test, AccuracyCheck) {
|
|
ACMRandom rnd(ACMRandom::DeterministicSeed());
|
|
uint32_t max_error = 0;
|
|
int64_t total_error = 0;
|
|
const int count_test_block = 1000;
|
|
DECLARE_ALIGNED_ARRAY(16, int16_t, test_input_block, kNumCoeffs);
|
|
DECLARE_ALIGNED_ARRAY(16, int16_t, test_temp_block, kNumCoeffs);
|
|
DECLARE_ALIGNED_ARRAY(16, uint8_t, dst, kNumCoeffs);
|
|
DECLARE_ALIGNED_ARRAY(16, uint8_t, src, kNumCoeffs);
|
|
|
|
for (int i = 0; i < count_test_block; ++i) {
|
|
// Initialize a test block with input range [-255, 255].
|
|
for (int j = 0; j < kNumCoeffs; ++j) {
|
|
src[j] = rnd.Rand8();
|
|
dst[j] = rnd.Rand8();
|
|
test_input_block[j] = src[j] - dst[j];
|
|
}
|
|
|
|
REGISTER_STATE_CHECK(fwd_txfm_(test_input_block, test_temp_block, 32));
|
|
REGISTER_STATE_CHECK(inv_txfm_(test_temp_block, dst, 32));
|
|
|
|
for (int j = 0; j < kNumCoeffs; ++j) {
|
|
const uint32_t diff = dst[j] - src[j];
|
|
const uint32_t error = diff * diff;
|
|
if (max_error < error)
|
|
max_error = error;
|
|
total_error += error;
|
|
}
|
|
}
|
|
|
|
if (version_ == 1) {
|
|
max_error /= 2;
|
|
total_error /= 45;
|
|
}
|
|
|
|
EXPECT_GE(1u, max_error)
|
|
<< "Error: 32x32 FDCT/IDCT has an individual round-trip error > 1";
|
|
|
|
EXPECT_GE(count_test_block, total_error)
|
|
<< "Error: 32x32 FDCT/IDCT has average round-trip error > 1 per block";
|
|
}
|
|
|
|
TEST_P(Trans32x32Test, CoeffCheck) {
|
|
ACMRandom rnd(ACMRandom::DeterministicSeed());
|
|
const int count_test_block = 1000;
|
|
|
|
DECLARE_ALIGNED_ARRAY(16, int16_t, input_block, kNumCoeffs);
|
|
DECLARE_ALIGNED_ARRAY(16, int16_t, output_ref_block, kNumCoeffs);
|
|
DECLARE_ALIGNED_ARRAY(16, int16_t, output_block, kNumCoeffs);
|
|
|
|
for (int i = 0; i < count_test_block; ++i) {
|
|
for (int j = 0; j < kNumCoeffs; ++j)
|
|
input_block[j] = rnd.Rand8() - rnd.Rand8();
|
|
|
|
const int stride = 32;
|
|
vp9_fdct32x32_c(input_block, output_ref_block, stride);
|
|
REGISTER_STATE_CHECK(fwd_txfm_(input_block, output_block, stride));
|
|
|
|
if (version_ == 0) {
|
|
for (int j = 0; j < kNumCoeffs; ++j)
|
|
EXPECT_EQ(output_block[j], output_ref_block[j])
|
|
<< "Error: 32x32 FDCT versions have mismatched coefficients";
|
|
} else {
|
|
for (int j = 0; j < kNumCoeffs; ++j)
|
|
EXPECT_GE(6, abs(output_block[j] - output_ref_block[j]))
|
|
<< "Error: 32x32 FDCT rd has mismatched coefficients";
|
|
}
|
|
}
|
|
}
|
|
|
|
TEST_P(Trans32x32Test, MemCheck) {
|
|
ACMRandom rnd(ACMRandom::DeterministicSeed());
|
|
const int count_test_block = 2000;
|
|
|
|
DECLARE_ALIGNED_ARRAY(16, int16_t, input_block, kNumCoeffs);
|
|
DECLARE_ALIGNED_ARRAY(16, int16_t, input_extreme_block, kNumCoeffs);
|
|
DECLARE_ALIGNED_ARRAY(16, int16_t, output_ref_block, kNumCoeffs);
|
|
DECLARE_ALIGNED_ARRAY(16, int16_t, output_block, kNumCoeffs);
|
|
|
|
for (int i = 0; i < count_test_block; ++i) {
|
|
// Initialize a test block with input range [-255, 255].
|
|
for (int j = 0; j < kNumCoeffs; ++j) {
|
|
input_block[j] = rnd.Rand8() - rnd.Rand8();
|
|
input_extreme_block[j] = rnd.Rand8() & 1 ? 255 : -255;
|
|
}
|
|
if (i == 0) {
|
|
for (int j = 0; j < kNumCoeffs; ++j)
|
|
input_extreme_block[j] = 255;
|
|
} else if (i == 1) {
|
|
for (int j = 0; j < kNumCoeffs; ++j)
|
|
input_extreme_block[j] = -255;
|
|
}
|
|
|
|
const int stride = 32;
|
|
vp9_fdct32x32_c(input_extreme_block, output_ref_block, stride);
|
|
REGISTER_STATE_CHECK(fwd_txfm_(input_extreme_block, output_block, stride));
|
|
|
|
// The minimum quant value is 4.
|
|
for (int j = 0; j < kNumCoeffs; ++j) {
|
|
if (version_ == 0) {
|
|
EXPECT_EQ(output_block[j], output_ref_block[j])
|
|
<< "Error: 32x32 FDCT versions have mismatched coefficients";
|
|
} else {
|
|
EXPECT_GE(6, abs(output_block[j] - output_ref_block[j]))
|
|
<< "Error: 32x32 FDCT rd has mismatched coefficients";
|
|
}
|
|
EXPECT_GE(4 * DCT_MAX_VALUE, abs(output_ref_block[j]))
|
|
<< "Error: 32x32 FDCT C has coefficient larger than 4*DCT_MAX_VALUE";
|
|
EXPECT_GE(4 * DCT_MAX_VALUE, abs(output_block[j]))
|
|
<< "Error: 32x32 FDCT has coefficient larger than "
|
|
<< "4*DCT_MAX_VALUE";
|
|
}
|
|
}
|
|
}
|
|
|
|
TEST_P(Trans32x32Test, InverseAccuracy) {
|
|
ACMRandom rnd(ACMRandom::DeterministicSeed());
|
|
const int count_test_block = 1000;
|
|
DECLARE_ALIGNED_ARRAY(16, int16_t, in, kNumCoeffs);
|
|
DECLARE_ALIGNED_ARRAY(16, int16_t, coeff, kNumCoeffs);
|
|
DECLARE_ALIGNED_ARRAY(16, uint8_t, dst, kNumCoeffs);
|
|
DECLARE_ALIGNED_ARRAY(16, uint8_t, src, kNumCoeffs);
|
|
|
|
for (int i = 0; i < count_test_block; ++i) {
|
|
double out_r[kNumCoeffs];
|
|
|
|
// Initialize a test block with input range [-255, 255]
|
|
for (int j = 0; j < kNumCoeffs; ++j) {
|
|
src[j] = rnd.Rand8();
|
|
dst[j] = rnd.Rand8();
|
|
in[j] = src[j] - dst[j];
|
|
}
|
|
|
|
reference_32x32_dct_2d(in, out_r);
|
|
for (int j = 0; j < kNumCoeffs; ++j)
|
|
coeff[j] = round(out_r[j]);
|
|
REGISTER_STATE_CHECK(inv_txfm_(coeff, dst, 32));
|
|
for (int j = 0; j < kNumCoeffs; ++j) {
|
|
const int diff = dst[j] - src[j];
|
|
const int error = diff * diff;
|
|
EXPECT_GE(1, error)
|
|
<< "Error: 32x32 IDCT has error " << error
|
|
<< " at index " << j;
|
|
}
|
|
}
|
|
}
|
|
|
|
using std::tr1::make_tuple;
|
|
|
|
INSTANTIATE_TEST_CASE_P(
|
|
C, Trans32x32Test,
|
|
::testing::Values(
|
|
make_tuple(&vp9_fdct32x32_c, &vp9_idct32x32_1024_add_c, 0),
|
|
make_tuple(&vp9_fdct32x32_rd_c, &vp9_idct32x32_1024_add_c, 1)));
|
|
|
|
#if HAVE_NEON_ASM
|
|
INSTANTIATE_TEST_CASE_P(
|
|
NEON, Trans32x32Test,
|
|
::testing::Values(
|
|
make_tuple(&vp9_fdct32x32_c,
|
|
&vp9_idct32x32_1024_add_neon, 0),
|
|
make_tuple(&vp9_fdct32x32_rd_c,
|
|
&vp9_idct32x32_1024_add_neon, 1)));
|
|
#endif
|
|
|
|
#if HAVE_SSE2
|
|
INSTANTIATE_TEST_CASE_P(
|
|
SSE2, Trans32x32Test,
|
|
::testing::Values(
|
|
make_tuple(&vp9_fdct32x32_sse2,
|
|
&vp9_idct32x32_1024_add_sse2, 0),
|
|
make_tuple(&vp9_fdct32x32_rd_sse2,
|
|
&vp9_idct32x32_1024_add_sse2, 1)));
|
|
#endif
|
|
|
|
#if HAVE_AVX2
|
|
INSTANTIATE_TEST_CASE_P(
|
|
AVX2, Trans32x32Test,
|
|
::testing::Values(
|
|
make_tuple(&vp9_fdct32x32_avx2,
|
|
&vp9_idct32x32_1024_add_sse2, 0),
|
|
make_tuple(&vp9_fdct32x32_rd_avx2,
|
|
&vp9_idct32x32_1024_add_sse2, 1)));
|
|
#endif
|
|
} // namespace
|