Skip to content

GitLab

Commit 8948fad3 authored by Simon Morlat's avatar Simon Morlat
Browse files

re-indent file with tabs. Everyone must use tabs in this company.

parent 00e7e778
Hide whitespace changes
Inline Side-by-side
Showing with 196 additions and 191 deletions
tester/mediastreamer2_neon_tester.c
View file @ 8948fad3
......@@ -28,15 +28,15 @@
#include <arm_neon.h>
static int tester_init() {
ortp_set_log_level_mask(ORTP_MESSAGE | ORTP_WARNING | ORTP_ERROR | ORTP_FATAL);
ms_init();
srand(time(0));
return 0;
ortp_set_log_level_mask(ORTP_MESSAGE | ORTP_WARNING | ORTP_ERROR | ORTP_FATAL);
ms_init();
srand(time(0));
return 0;
}
static int tester_cleanup() {
ms_exit();
return 0;
ms_exit();
return 0;
}
......@@ -45,150 +45,155 @@ extern int libspeex_cpu_features;
extern spx_int32_t inner_prod( const spx_int16_t* a, const spx_int16_t* b, int len);
#if 0 /* These are inner product implementations that we used for fixing speex's one */
spx_int32_t ms_inner_product_neon_xcode(const spx_int16_t *a, const spx_int16_t *b, unsigned int len){
int16x4_t a4;
int16x4_t b4;
spx_int32_t sum = 0; // sum is 0
int32x4_t part = vdupq_n_s32(0);
int32x2_t dAccL, dAccH;
spx_int32_t s;
len >>=2;
while (len--) {
a4 = vld1_s16(a); // load 4 * 16b from a
b4 = vld1_s16(b); // load 4 * 16b from b
part = vmlal_s16(part, a4, b4); // part[n] += a[n] * b[n]
a+=4; b+=4;
}
/* Split 128-bit qAccumulator into 64-bit dAccL and dAccH for
* accumulation */
dAccL = vget_low_s32(part);
dAccH = vget_high_s32(part);
/* Accumulate 2 lanes in dAccL and dAccH into 2 lanes in dAccL */
dAccL = vadd_s32(dAccL, dAccH);
/* Accumulate 2 lanes in dAccL into first (and second) lane of dAccL */
dAccL = vpadd_s32(dAccL, dAccL);
s = vget_lane_s32(dAccL, 0);
// ms_message("S[%d] = %d", len,s);
/* Add accumulated value to retval */
sum += (s >> 6);
return sum;
spx_int32_t ms_inner_product_neon_xcode(const spx_int16_t *a, const spx_int16_t *b, unsigned int len) {
int16x4_t a4;
int16x4_t b4;
spx_int32_t sum = 0; // sum is 0
int32x4_t part = vdupq_n_s32(0);
int32x2_t dAccL, dAccH;
spx_int32_t s;
len >>=2;
while (len--) {
a4 = vld1_s16(a); // load 4 * 16b from a
b4 = vld1_s16(b); // load 4 * 16b from b
part = vmlal_s16(part, a4, b4); // part[n] += a[n] * b[n]
a+=4;
b+=4;
}
/* Split 128-bit qAccumulator into 64-bit dAccL and dAccH for
* accumulation */
dAccL = vget_low_s32(part);
dAccH = vget_high_s32(part);
/* Accumulate 2 lanes in dAccL and dAccH into 2 lanes in dAccL */
dAccL = vadd_s32(dAccL, dAccH);
/* Accumulate 2 lanes in dAccL into first (and second) lane of dAccL */
dAccL = vpadd_s32(dAccL, dAccL);
s = vget_lane_s32(dAccL, 0);
// ms_message("S[%d] = %d", len,s);
/* Add accumulated value to retval */
sum += (s >> 6);
return sum;
}
spx_int32_t ms_inner_product_neon_xcode_optim(const spx_int16_t *a, const spx_int16_t *b, unsigned int len){
int16x4_t a4;
int16x4_t b4;
spx_int32_t sum = 0; // sum is 0
int32x4_t partial = vdupq_n_s32(0);
int64x2_t back;
spx_int32_t ms_inner_product_neon_xcode_optim(const spx_int16_t *a, const spx_int16_t *b, unsigned int len) {
int16x4_t a4;
int16x4_t b4;
spx_int32_t sum = 0; // sum is 0
int32x4_t partial = vdupq_n_s32(0);
int64x2_t back;
len >>=2;
len >>=2;
while (len--) {
a4 = vld1_s16(a); // load 4 * 16b from a
b4 = vld1_s16(b); // load 4 * 16b from b
while (len--) {
a4 = vld1_s16(a); // load 4 * 16b from a
b4 = vld1_s16(b); // load 4 * 16b from b
partial = vmlal_s16(partial, a4, b4); // part[n] += a[n] * b[n] vector multiply and add
a+=4; b+=4;
}
back = vpaddlq_s32(partial); // sum the 4 s32 in 2 64b
back = vshrq_n_s64(back, 6); // shift by 6
partial = vmlal_s16(partial, a4, b4); // part[n] += a[n] * b[n] vector multiply and add
a+=4;
b+=4;
}
back = vpaddlq_s32(partial); // sum the 4 s32 in 2 64b
back = vshrq_n_s64(back, 6); // shift by 6
sum += vgetq_lane_s64(back, 0) + vgetq_lane_s64(back, 1);
partial = vdupq_n_s32(0); // reset partial sum
sum += vgetq_lane_s64(back, 0) + vgetq_lane_s64(back, 1);
partial = vdupq_n_s32(0); // reset partial sum
return sum;
return sum;
}
spx_int32_t ms_inner_product_neon_xcode_optim8(const spx_int16_t *a, const spx_int16_t *b, unsigned int len){
int16x8_t a8;
int16x8_t b8;
spx_int32_t sum = 0; // sum is 0
int32x4_t partial = vdupq_n_s32(0);
int64x2_t back;
spx_int32_t ms_inner_product_neon_xcode_optim8(const spx_int16_t *a, const spx_int16_t *b, unsigned int len) {
int16x8_t a8;
int16x8_t b8;
spx_int32_t sum = 0; // sum is 0
int32x4_t partial = vdupq_n_s32(0);
int64x2_t back;
len >>=3;
len >>=3;
while (len--) {
a8 = vld1q_s16(a); // load 8 16b from a
b8 = vld1q_s16(b); // load 8 16b from b
while (len--) {
a8 = vld1q_s16(a); // load 8 16b from a
b8 = vld1q_s16(b); // load 8 16b from b
partial = vmlal_s16(partial, vget_low_s16(a8), vget_low_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
partial = vmlal_s16(partial, vget_high_s16(a8), vget_high_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
a+=8;
b+=8;
}
back = vpaddlq_s32(partial); // sum the 4 s32 in 2 64b
back = vshrq_n_s64(back, 6); // shift by 6
partial = vmlal_s16(partial, vget_low_s16(a8), vget_low_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
partial = vmlal_s16(partial, vget_high_s16(a8), vget_high_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
a+=8; b+=8;
}
back = vpaddlq_s32(partial); // sum the 4 s32 in 2 64b
back = vshrq_n_s64(back, 6); // shift by 6
sum += vgetq_lane_s64(back, 0) + vgetq_lane_s64(back, 1);
partial = vdupq_n_s32(0); // reset partial sum
sum += vgetq_lane_s64(back, 0) + vgetq_lane_s64(back, 1);
partial = vdupq_n_s32(0); // reset partial sum
return sum;
return sum;
}
spx_int32_t ms_inner_product_neon_xcode_optim16(const spx_int16_t *a, const spx_int16_t *b, unsigned int len){
int16x8_t a8,b8;
int16x8_t c8,d8;
spx_int32_t sum = 0; // sum is 0
int32x4_t partial = vdupq_n_s32(0);
int32x4_t p2 = vdupq_n_s32(0);
int64x2_t back;
len >>=4;
while (len--) {
a8 = vld1q_s16(a); // load 8 16b from a
b8 = vld1q_s16(b); // load 8 16b from b
c8 = vld1q_s16(a+8); // load 8 16b from b
d8 = vld1q_s16(b+8); // load 8 16b from b
partial = vmlal_s16(partial, vget_low_s16(a8), vget_low_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
partial = vmlal_s16(partial, vget_high_s16(a8), vget_high_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
p2 = vmlal_s16(p2, vget_low_s16(c8), vget_low_s16(d8)); // part[n] += a[n] * b[n] vector multiply and add
p2 = vmlal_s16(p2, vget_high_s16(c8), vget_high_s16(d8)); // part[n] += a[n] * b[n] vector multiply and add
a+=16; b+=16;
}
back = vpaddlq_s32(partial); // sum the 4 s32 in 2 64b
back = vpadalq_s32(back, p2); // add to all these the 2nd partial
back = vshrq_n_s64(back, 6); // shift by 6
sum += vgetq_lane_s64(back, 0) + vgetq_lane_s64(back, 1);
partial = vdupq_n_s32(0); // reset partial sum
return sum;
spx_int32_t ms_inner_product_neon_xcode_optim16(const spx_int16_t *a, const spx_int16_t *b, unsigned int len) {
int16x8_t a8,b8;
int16x8_t c8,d8;
spx_int32_t sum = 0; // sum is 0
int32x4_t partial = vdupq_n_s32(0);
int32x4_t p2 = vdupq_n_s32(0);
int64x2_t back;
len >>=4;
while (len--) {
a8 = vld1q_s16(a); // load 8 16b from a
b8 = vld1q_s16(b); // load 8 16b from b
c8 = vld1q_s16(a+8); // load 8 16b from b
d8 = vld1q_s16(b+8); // load 8 16b from b
partial = vmlal_s16(partial, vget_low_s16(a8), vget_low_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
partial = vmlal_s16(partial, vget_high_s16(a8), vget_high_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
p2 = vmlal_s16(p2, vget_low_s16(c8), vget_low_s16(d8)); // part[n] += a[n] * b[n] vector multiply and add
p2 = vmlal_s16(p2, vget_high_s16(c8), vget_high_s16(d8)); // part[n] += a[n] * b[n] vector multiply and add
a+=16;
b+=16;
}
back = vpaddlq_s32(partial); // sum the 4 s32 in 2 64b
back = vpadalq_s32(back, p2); // add to all these the 2nd partial
back = vshrq_n_s64(back, 6); // shift by 6
sum += vgetq_lane_s64(back, 0) + vgetq_lane_s64(back, 1);
partial = vdupq_n_s32(0); // reset partial sum
return sum;
}
/* intrinsics */
spx_int32_t ms_inner_product_neon_intrinsics(const spx_int16_t *a, const spx_int16_t *b, unsigned int len){
int16x8_t a8;
int16x8_t b8;
int32x4_t partial = vdupq_n_s32(0);
int64x2_t back;
len >>=3;
while (len--) {
a8 = vld1q_s16(a); // load 8 16b from a
b8 = vld1q_s16(b); // load 8 16b from b
partial = vmlal_s16(partial, vget_low_s16(a8), vget_low_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
partial = vmlal_s16(partial, vget_high_s16(a8), vget_high_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
a+=8; b+=8;
}
back = vpaddlq_s32(partial); // sum the 4 s32 in 2 64b
back = vshrq_n_s64(back, 6); // shift by 6
return vgetq_lane_s64(back, 0) + vgetq_lane_s64(back, 1);
spx_int32_t ms_inner_product_neon_intrinsics(const spx_int16_t *a, const spx_int16_t *b, unsigned int len) {
int16x8_t a8;
int16x8_t b8;
int32x4_t partial = vdupq_n_s32(0);
int64x2_t back;
len >>=3;
while (len--) {
a8 = vld1q_s16(a); // load 8 16b from a
b8 = vld1q_s16(b); // load 8 16b from b
partial = vmlal_s16(partial, vget_low_s16(a8), vget_low_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
partial = vmlal_s16(partial, vget_high_s16(a8), vget_high_s16(b8)); // part[n] += a[n] * b[n] vector multiply and add
a+=8;
b+=8;
}
back = vpaddlq_s32(partial); // sum the 4 s32 in 2 64b
back = vshrq_n_s64(back, 6); // shift by 6
return vgetq_lane_s64(back, 0) + vgetq_lane_s64(back, 1);
}
#endif
......@@ -197,76 +202,76 @@ static void inner_product_test(void) {
#ifdef SPEEX_LIB_CPU_FEATURE_NEON
#define SAMPLE_SIZE 64 /* has to be %8 and < 64 ! */
#define ITERATIONS 1000000
static spx_int16_t test_sample[SAMPLE_SIZE];
static spx_int16_t test_sample2[SAMPLE_SIZE];
int length = SAMPLE_SIZE;
uint64_t soft_ms, neon_ms;
int i;
volatile spx_int32_t non_neon_result;
volatile spx_int32_t neon_result;
float percent_off;
bool_t fast_enough;
// put some values to process
for( i = 0; i<SAMPLE_SIZE; i++){
test_sample[i] = ortp_random() % 16384;
test_sample2[i] = ortp_random() % 16384;
}
// ARM_NEON is enabled, we can force libspeex to use it
libspeex_cpu_features = SPEEX_LIB_CPU_FEATURE_NEON;
// disable neon & perform inner product
libspeex_cpu_features &= ~SPEEX_LIB_CPU_FEATURE_NEON;
i = ITERATIONS;
{
uint64_t start = ms_get_cur_time_ms();
while (i--) {
non_neon_result = inner_prod((const spx_int16_t*)test_sample, (const spx_int16_t*)test_sample2, length);
}
soft_ms = ms_get_cur_time_ms() - start;
}
// enable neon and perform the same operation
libspeex_cpu_features |= SPEEX_LIB_CPU_FEATURE_NEON;
i = ITERATIONS;
{
uint64_t start = ms_get_cur_time_ms();
while (i--) {
neon_result= inner_prod((const spx_int16_t*)test_sample, (const spx_int16_t*)test_sample2, length);
}
neon_ms = ms_get_cur_time_ms() - start;
}
percent_off = ((float)abs(non_neon_result-neon_result))/MAX(non_neon_result, neon_result)*100;
ms_debug("%10d, NON Neon: %10d - diff: %d - percent off: %f",
non_neon_result, neon_result, abs(non_neon_result-neon_result), percent_off);
fast_enough = (float)neon_ms < (float)soft_ms/5;
// we expect the result to be very similar and at least 5 times faster with NEON
CU_ASSERT(percent_off < 1.0);
CU_ASSERT(fast_enough);
ms_message("NEON = %llu ms, SOFT: %llu ms", neon_ms, soft_ms);
if( !fast_enough ){
ms_error("NEON not fast enough it seems");
}
static spx_int16_t test_sample[SAMPLE_SIZE];
static spx_int16_t test_sample2[SAMPLE_SIZE];
int length = SAMPLE_SIZE;
uint64_t soft_ms, neon_ms;
int i;
volatile spx_int32_t non_neon_result;
volatile spx_int32_t neon_result;
float percent_off;
bool_t fast_enough;
// put some values to process
for( i = 0; i<SAMPLE_SIZE; i++) {
test_sample[i] = ortp_random() % 16384;
test_sample2[i] = ortp_random() % 16384;
}
// ARM_NEON is enabled, we can force libspeex to use it
libspeex_cpu_features = SPEEX_LIB_CPU_FEATURE_NEON;
// disable neon & perform inner product
libspeex_cpu_features &= ~SPEEX_LIB_CPU_FEATURE_NEON;
i = ITERATIONS;
{
uint64_t start = ms_get_cur_time_ms();
while (i--) {
non_neon_result = inner_prod((const spx_int16_t*)test_sample, (const spx_int16_t*)test_sample2, length);
}
soft_ms = ms_get_cur_time_ms() - start;
}
// enable neon and perform the same operation
libspeex_cpu_features |= SPEEX_LIB_CPU_FEATURE_NEON;
i = ITERATIONS;
{
uint64_t start = ms_get_cur_time_ms();
while (i--) {
neon_result= inner_prod((const spx_int16_t*)test_sample, (const spx_int16_t*)test_sample2, length);
}
neon_ms = ms_get_cur_time_ms() - start;
}
percent_off = ((float)abs(non_neon_result-neon_result))/MAX(non_neon_result, neon_result)*100;
ms_debug("%10d, NON Neon: %10d - diff: %d - percent off: %f",
non_neon_result, neon_result, abs(non_neon_result-neon_result), percent_off);
fast_enough = (float)neon_ms < (float)soft_ms/5;
// we expect the result to be very similar and at least 5 times faster with NEON
CU_ASSERT(percent_off < 1.0);
CU_ASSERT(fast_enough);
ms_message("NEON = %llu ms, SOFT: %llu ms", neon_ms, soft_ms);
if( !fast_enough ) {
ms_error("NEON not fast enough it seems");
}
#else
ms_warning("Test skipped, not using the speex from git://git.linphone.org/speex.git with NEON support");
ms_warning("Test skipped, not using the speex from git://git.linphone.org/speex.git with NEON support");
#endif
}
static test_t tests[] = {
{ "Inner product", inner_product_test }
{ "Inner product", inner_product_test }
};
test_suite_t neon_test_suite = {
"NEON",
tester_init,
tester_cleanup,
sizeof(tests)/sizeof(test_t),
tests
"NEON",
tester_init,
tester_cleanup,
sizeof(tests)/sizeof(test_t),
tests
};
#endif // ARM NEON
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment