diff options
author | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-07 18:28:17 +0000 |
---|---|---|
committer | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-07 18:28:17 +0000 |
commit | 7a46c07230b8d8108c0e8e80df4522d0ac116538 (patch) | |
tree | d483300dab478b994fe199a5d19d18d74153718a /spa/plugins/audioconvert/benchmark-fmt-ops.c | |
parent | Initial commit. (diff) | |
download | pipewire-upstream/0.3.65.tar.xz pipewire-upstream/0.3.65.zip |
Adding upstream version 0.3.65.upstream/0.3.65upstream
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'spa/plugins/audioconvert/benchmark-fmt-ops.c')
-rw-r--r-- | spa/plugins/audioconvert/benchmark-fmt-ops.c | 323 |
1 files changed, 323 insertions, 0 deletions
diff --git a/spa/plugins/audioconvert/benchmark-fmt-ops.c b/spa/plugins/audioconvert/benchmark-fmt-ops.c new file mode 100644 index 0000000..2a0d4e8 --- /dev/null +++ b/spa/plugins/audioconvert/benchmark-fmt-ops.c @@ -0,0 +1,323 @@ +/* Spa + * + * Copyright © 2019 Wim Taymans + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + * DEALINGS IN THE SOFTWARE. + */ + +#include "config.h" + +#include <string.h> +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <errno.h> +#include <time.h> + +#include "test-helper.h" +#include "fmt-ops.h" + +static uint32_t cpu_flags; + +typedef void (*convert_func_t) (struct convert *conv, void * SPA_RESTRICT dst[], + const void * SPA_RESTRICT src[], uint32_t n_samples); + +struct stats { + uint32_t n_samples; + uint32_t n_channels; + uint64_t perf; + const char *name; + const char *impl; +}; + +#define MAX_SAMPLES 4096 +#define MAX_CHANNELS 11 + +#define MAX_COUNT 100 + +static uint8_t samp_in[MAX_SAMPLES * MAX_CHANNELS * 4]; +static uint8_t samp_out[MAX_SAMPLES * MAX_CHANNELS * 4]; + +static const int sample_sizes[] = { 0, 1, 128, 513, 4096 }; +static const int channel_counts[] = { 1, 2, 4, 6, 8, 11 }; + +#define MAX_RESULTS SPA_N_ELEMENTS(sample_sizes) * SPA_N_ELEMENTS(channel_counts) * 70 + +static uint32_t n_results = 0; +static struct stats results[MAX_RESULTS]; + +static void run_test1(const char *name, const char *impl, bool in_packed, bool out_packed, + convert_func_t func, int n_channels, int n_samples) +{ + int i, j; + const void *ip[n_channels]; + void *op[n_channels]; + struct timespec ts; + uint64_t count, t1, t2; + struct convert conv; + + conv.n_channels = n_channels; + + for (j = 0; j < n_channels; j++) { + ip[j] = &samp_in[j * n_samples * 4]; + op[j] = &samp_out[j * n_samples * 4]; + } + + clock_gettime(CLOCK_MONOTONIC, &ts); + t1 = SPA_TIMESPEC_TO_NSEC(&ts); + + count = 0; + for (i = 0; i < MAX_COUNT; i++) { + func(&conv, op, ip, n_samples); + count++; + } + clock_gettime(CLOCK_MONOTONIC, &ts); + t2 = SPA_TIMESPEC_TO_NSEC(&ts); + + spa_assert(n_results < MAX_RESULTS); + + results[n_results++] = (struct stats) { + .n_samples = n_samples, + .n_channels = n_channels, + .perf = count * (uint64_t)SPA_NSEC_PER_SEC / (t2 - t1), + .name = name, + .impl = impl + }; +} + +static void run_testc(const char *name, const char *impl, bool in_packed, bool out_packed, convert_func_t func, + int channel_count) +{ + SPA_FOR_EACH_ELEMENT_VAR(sample_sizes, s) { + run_test1(name, impl, in_packed, out_packed, func, channel_count, + (*s + (channel_count -1)) / channel_count); + } +} + +static void run_test(const char *name, const char *impl, bool in_packed, bool out_packed, convert_func_t func) +{ + SPA_FOR_EACH_ELEMENT_VAR(sample_sizes, s) { + SPA_FOR_EACH_ELEMENT_VAR(channel_counts, c) { + run_test1(name, impl, in_packed, out_packed, func, *c, (*s + (*c -1)) / *c); + } + } +} + +static void test_f32_u8(void) +{ + run_test("test_f32_u8", "c", true, true, conv_f32_to_u8_c); + run_test("test_f32d_u8", "c", false, true, conv_f32d_to_u8_c); + run_test("test_f32_u8d", "c", true, false, conv_f32_to_u8d_c); + run_test("test_f32d_u8d", "c", false, false, conv_f32d_to_u8d_c); +} + +static void test_u8_f32(void) +{ + run_test("test_u8_f32", "c", true, true, conv_u8_to_f32_c); + run_test("test_u8d_f32", "c", false, true, conv_u8d_to_f32_c); + run_test("test_u8_f32d", "c", true, false, conv_u8_to_f32d_c); + run_test("test_u8d_f32d", "c", false, false, conv_u8d_to_f32d_c); +} + +static void test_f32_s16(void) +{ + run_test("test_f32_s16", "c", true, true, conv_f32_to_s16_c); + run_test("test_f32d_s16", "c", false, true, conv_f32d_to_s16_c); +#if defined (HAVE_SSE2) + if (cpu_flags & SPA_CPU_FLAG_SSE2) { + run_test("test_f32d_s16", "sse2", false, true, conv_f32d_to_s16_sse2); + run_testc("test_f32d_s16_2", "sse2", false, true, conv_f32d_to_s16_2_sse2, 2); + } +#endif +#if defined (HAVE_AVX2) + if (cpu_flags & SPA_CPU_FLAG_AVX2) { + run_test("test_f32d_s16", "avx2", false, true, conv_f32d_to_s16_avx2); + run_testc("test_f32d_s16_2", "avx2", false, true, conv_f32d_to_s16_2_avx2, 2); + run_testc("test_f32d_s16_4", "avx2", false, true, conv_f32d_to_s16_4_avx2, 4); + } +#endif + run_test("test_f32_s16d", "c", true, false, conv_f32_to_s16d_c); + run_test("test_f32d_s16d", "c", false, false, conv_f32d_to_s16d_c); +} + +static void test_s16_f32(void) +{ + run_test("test_s16_f32", "c", true, true, conv_s16_to_f32_c); + run_test("test_s16d_f32", "c", false, true, conv_s16d_to_f32_c); + run_test("test_s16_f32d", "c", true, false, conv_s16_to_f32d_c); +#if defined (HAVE_SSE2) + if (cpu_flags & SPA_CPU_FLAG_SSE2) { + run_test("test_s16_f32d", "sse2", true, false, conv_s16_to_f32d_sse2); + run_testc("test_s16_f32d_2", "sse2", true, false, conv_s16_to_f32d_2_sse2, 2); + } +#endif +#if defined (HAVE_AVX2) + if (cpu_flags & SPA_CPU_FLAG_AVX2) { + run_test("test_s16_f32d", "avx2", true, false, conv_s16_to_f32d_avx2); + run_testc("test_s16_f32d_2", "avx2", true, false, conv_s16_to_f32d_2_avx2, 2); + } +#endif + run_test("test_s16d_f32d", "c", false, false, conv_s16d_to_f32d_c); +} + +static void test_f32_s32(void) +{ + run_test("test_f32_s32", "c", true, true, conv_f32_to_s32_c); + run_test("test_f32d_s32", "c", false, true, conv_f32d_to_s32_c); +#if defined (HAVE_SSE2) + if (cpu_flags & SPA_CPU_FLAG_SSE2) { + run_test("test_f32d_s32", "sse2", false, true, conv_f32d_to_s32_sse2); + } +#endif +#if defined (HAVE_AVX2) + if (cpu_flags & SPA_CPU_FLAG_AVX2) { + run_test("test_f32d_s32", "avx2", false, true, conv_f32d_to_s32_avx2); + } +#endif + run_test("test_f32_s32d", "c", true, false, conv_f32_to_s32d_c); + run_test("test_f32d_s32d", "c", false, false, conv_f32d_to_s32d_c); +} + +static void test_s32_f32(void) +{ + run_test("test_s32_f32", "c", true, true, conv_s32_to_f32_c); + run_test("test_s32d_f32", "c", false, true, conv_s32d_to_f32_c); +#if defined (HAVE_SSE2) + if (cpu_flags & SPA_CPU_FLAG_SSE2) { + run_test("test_s32_f32d", "sse2", true, false, conv_s32_to_f32d_sse2); + } +#endif +#if defined (HAVE_AVX2) + if (cpu_flags & SPA_CPU_FLAG_AVX2) { + run_test("test_s32_f32d", "avx2", true, false, conv_s32_to_f32d_avx2); + } +#endif + run_test("test_s32_f32d", "c", true, false, conv_s32_to_f32d_c); + run_test("test_s32d_f32d", "c", false, false, conv_s32d_to_f32d_c); +} + +static void test_f32_s24(void) +{ + run_test("test_f32_s24", "c", true, true, conv_f32_to_s24_c); + run_test("test_f32d_s24", "c", false, true, conv_f32d_to_s24_c); + run_test("test_f32_s24d", "c", true, false, conv_f32_to_s24d_c); + run_test("test_f32d_s24d", "c", false, false, conv_f32d_to_s24d_c); +} + +static void test_s24_f32(void) +{ + run_test("test_s24_f32", "c", true, true, conv_s24_to_f32_c); + run_test("test_s24d_f32", "c", false, true, conv_s24d_to_f32_c); + run_test("test_s24_f32d", "c", true, false, conv_s24_to_f32d_c); +#if defined (HAVE_SSE2) + if (cpu_flags & SPA_CPU_FLAG_SSE2) { + run_test("test_s24_f32d", "sse2", true, false, conv_s24_to_f32d_sse2); + } +#endif +#if defined (HAVE_AVX2) + if (cpu_flags & SPA_CPU_FLAG_AVX2) { + run_test("test_s24_f32d", "avx2", true, false, conv_s24_to_f32d_avx2); + } +#endif +#if defined (HAVE_SSSE3) + if (cpu_flags & SPA_CPU_FLAG_SSSE3) { + run_test("test_s24_f32d", "ssse3", true, false, conv_s24_to_f32d_ssse3); + } +#endif +#if defined (HAVE_SSE41) + if (cpu_flags & SPA_CPU_FLAG_SSE41) { + run_test("test_s24_f32d", "sse41", true, false, conv_s24_to_f32d_sse41); + } +#endif + run_test("test_s24d_f32d", "c", false, false, conv_s24d_to_f32d_c); +} + +static void test_f32_s24_32(void) +{ + run_test("test_f32_s24_32", "c", true, true, conv_f32_to_s24_32_c); + run_test("test_f32d_s24_32", "c", false, true, conv_f32d_to_s24_32_c); + run_test("test_f32_s24_32d", "c", true, false, conv_f32_to_s24_32d_c); + run_test("test_f32d_s24_32d", "c", false, false, conv_f32d_to_s24_32d_c); +} + +static void test_s24_32_f32(void) +{ + run_test("test_s24_32_f32", "c", true, true, conv_s24_32_to_f32_c); + run_test("test_s24_32d_f32", "c", false, true, conv_s24_32d_to_f32_c); + run_test("test_s24_32_f32d", "c", true, false, conv_s24_32_to_f32d_c); + run_test("test_s24_32d_f32d", "c", false, false, conv_s24_32d_to_f32d_c); +} + +static void test_interleave(void) +{ + run_test("test_8d_to_8", "c", false, true, conv_8d_to_8_c); + run_test("test_16d_to_16", "c", false, true, conv_16d_to_16_c); + run_test("test_24d_to_24", "c", false, true, conv_24d_to_24_c); + run_test("test_32d_to_32", "c", false, true, conv_32d_to_32_c); +} + +static void test_deinterleave(void) +{ + run_test("test_8_to_8d", "c", true, false, conv_8_to_8d_c); + run_test("test_16_to_16d", "c", true, false, conv_16_to_16d_c); + run_test("test_24_to_24d", "c", true, false, conv_24_to_24d_c); + run_test("test_32_to_32d", "c", true, false, conv_32_to_32d_c); +} + +static int compare_func(const void *_a, const void *_b) +{ + const struct stats *a = _a, *b = _b; + int diff; + if ((diff = strcmp(a->name, b->name)) != 0) return diff; + if ((diff = a->n_samples - b->n_samples) != 0) return diff; + if ((diff = a->n_channels - b->n_channels) != 0) return diff; + if ((diff = b->perf - a->perf) != 0) return diff; + return 0; +} + +int main(int argc, char *argv[]) +{ + uint32_t i; + + cpu_flags = get_cpu_flags(); + printf("got get CPU flags %d\n", cpu_flags); + + test_f32_u8(); + test_u8_f32(); + test_f32_s16(); + test_s16_f32(); + test_f32_s32(); + test_s32_f32(); + test_f32_s24(); + test_s24_f32(); + test_f32_s24_32(); + test_s24_32_f32(); + test_interleave(); + test_deinterleave(); + + qsort(results, n_results, sizeof(struct stats), compare_func); + + for (i = 0; i < n_results; i++) { + struct stats *s = &results[i]; + fprintf(stderr, "%-12."PRIu64" \t%-32.32s %s \t samples %d, channels %d\n", + s->perf, s->name, s->impl, s->n_samples, s->n_channels); + } + return 0; +} |