154 lines
5.4 KiB
C
154 lines
5.4 KiB
C
/* SPDX-License-Identifier: MIT
|
|
*
|
|
* 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 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.
|
|
*
|
|
* Copyright:
|
|
* 2023 Yi-Yen Chung <eric681@andestech.com> (Copyright owned by Andes Technology)
|
|
*/
|
|
|
|
#if !defined(SIMDE_ARM_NEON_MMLAQ_H)
|
|
#define SIMDE_ARM_NEON_MMLAQ_H
|
|
|
|
#include "types.h"
|
|
#include "cgt.h"
|
|
#include "bsl.h"
|
|
|
|
HEDLEY_DIAGNOSTIC_PUSH
|
|
SIMDE_DISABLE_UNWANTED_DIAGNOSTICS
|
|
SIMDE_BEGIN_DECLS_
|
|
|
|
SIMDE_FUNCTION_ATTRIBUTES
|
|
simde_int32x4_t
|
|
simde_vmmlaq_s32(simde_int32x4_t r, simde_int8x16_t a, simde_int8x16_t b) {
|
|
// I8MM is optional feature. src: https://patchwork.ffmpeg.org/project/ffmpeg/patch/20230530123043.52940-2-martin@martin.st/
|
|
#if defined(SIMDE_ARM_NEON_A32V8_NATIVE) && defined(SIMDE_ARCH_ARM_MATMUL_INT8)
|
|
return vmmlaq_s32(r, a, b);
|
|
#else
|
|
simde_int8x16_private
|
|
a_ = simde_int8x16_to_private(a),
|
|
b_ = simde_int8x16_to_private(b);
|
|
simde_int32x4_private
|
|
r_ = simde_int32x4_to_private(r),
|
|
ret;
|
|
|
|
for (size_t k = 0 ; k < (sizeof(ret.values) / sizeof(ret.values[0])) ; k++) {
|
|
ret.values[k] = r_.values[k];
|
|
for (size_t i = 0 ; i < (sizeof(a_.values) / sizeof(a_.values[0]) / 2) ; i++) {
|
|
ret.values[k] += a_.values[(k/2)*8+i] * b_.values[(k%2)*8+i];
|
|
}
|
|
}
|
|
return simde_int32x4_from_private(ret);
|
|
#endif
|
|
}
|
|
#if defined(SIMDE_ARM_NEON_A32V8_ENABLE_NATIVE_ALIASES)
|
|
#undef vmmlaq_s32
|
|
#define vmmlaq_s32(r, a, b) simde_vmmlaq_s32((r), (a), (b))
|
|
#endif
|
|
|
|
SIMDE_FUNCTION_ATTRIBUTES
|
|
simde_uint32x4_t
|
|
simde_vmmlaq_u32(simde_uint32x4_t r, simde_uint8x16_t a, simde_uint8x16_t b) {
|
|
#if defined(SIMDE_ARM_NEON_A32V8_NATIVE) && defined(SIMDE_ARCH_ARM_MATMUL_INT8)
|
|
return vmmlaq_u32(r, a, b);
|
|
#else
|
|
simde_uint8x16_private
|
|
a_ = simde_uint8x16_to_private(a),
|
|
b_ = simde_uint8x16_to_private(b);
|
|
simde_uint32x4_private
|
|
r_ = simde_uint32x4_to_private(r),
|
|
ret;
|
|
|
|
for (size_t k = 0 ; k < (sizeof(ret.values) / sizeof(ret.values[0])) ; k++) {
|
|
ret.values[k] = r_.values[k];
|
|
for (size_t i = 0 ; i < (sizeof(a_.values) / sizeof(a_.values[0]) / 2) ; i++) {
|
|
ret.values[k] += a_.values[(k/2)*8+i] * b_.values[(k%2)*8+i];
|
|
}
|
|
}
|
|
return simde_uint32x4_from_private(ret);
|
|
#endif
|
|
}
|
|
#if defined(SIMDE_ARM_NEON_A32V8_ENABLE_NATIVE_ALIASES)
|
|
#undef vmmlaq_u32
|
|
#define vmmlaq_u32(r, a, b) simde_vmmlaq_u32((r), (a), (b))
|
|
#endif
|
|
|
|
SIMDE_FUNCTION_ATTRIBUTES
|
|
simde_int32x4_t
|
|
simde_vusmmlaq_s32(simde_int32x4_t r, simde_uint8x16_t a, simde_int8x16_t b) {
|
|
#if defined(SIMDE_ARM_NEON_A32V8_NATIVE) && defined(SIMDE_ARCH_ARM_MATMUL_INT8)
|
|
return vusmmlaq_s32(r, a, b);
|
|
#else
|
|
simde_uint8x16_private
|
|
a_ = simde_uint8x16_to_private(a);
|
|
simde_int8x16_private
|
|
b_ = simde_int8x16_to_private(b);
|
|
simde_int32x4_private
|
|
r_ = simde_int32x4_to_private(r),
|
|
ret;
|
|
|
|
for (size_t k = 0 ; k < (sizeof(ret.values) / sizeof(ret.values[0])) ; k++) {
|
|
ret.values[k] = r_.values[k];
|
|
for (size_t i = 0 ; i < (sizeof(a_.values) / sizeof(a_.values[0]) / 2) ; i++) {
|
|
ret.values[k] += a_.values[(k/2)*8+i] * b_.values[(k%2)*8+i];
|
|
}
|
|
}
|
|
return simde_int32x4_from_private(ret);
|
|
#endif
|
|
}
|
|
#if defined(SIMDE_ARM_NEON_A32V8_ENABLE_NATIVE_ALIASES)
|
|
#undef vusmmlaq_s32
|
|
#define vusmmlaq_s32(r, a, b) simde_vusmmlaq_s32((r), (a), (b))
|
|
#endif
|
|
|
|
SIMDE_FUNCTION_ATTRIBUTES
|
|
simde_float32x4_t
|
|
simde_vbfmmlaq_f32(simde_float32x4_t r, simde_bfloat16x8_t a, simde_bfloat16x8_t b) {
|
|
#if defined(SIMDE_ARM_NEON_A32V8_NATIVE) && defined(SIMDE_ARCH_ARM_MATMUL_INT8) && \
|
|
defined(SIMDE_ARM_NEON_BF16)
|
|
return vbfmmlaq_f32(r, a, b);
|
|
#else
|
|
simde_bfloat16x8_private
|
|
a_ = simde_bfloat16x8_to_private(a),
|
|
b_ = simde_bfloat16x8_to_private(b);
|
|
simde_float32x4_private
|
|
r_ = simde_float32x4_to_private(r),
|
|
ret;
|
|
|
|
for (size_t k = 0 ; k < (sizeof(ret.values) / sizeof(ret.values[0])) ; k++) {
|
|
ret.values[k] = r_.values[k];
|
|
for (size_t i = 0 ; i < (sizeof(a_.values) / sizeof(a_.values[0]) / 2) ; i++) {
|
|
ret.values[k] += simde_bfloat16_to_float32(a_.values[(k/2)*4+i]) *
|
|
simde_bfloat16_to_float32(b_.values[(k%2)*4+i]);
|
|
}
|
|
}
|
|
return simde_float32x4_from_private(ret);
|
|
#endif
|
|
}
|
|
#if defined(SIMDE_ARM_NEON_A32V8_ENABLE_NATIVE_ALIASES)
|
|
#undef vbfmmlaq_f32
|
|
#define vbfmmlaq_f32(r, a, b) simde_vbfmmlaq_f32((r), (a), (b))
|
|
#endif
|
|
|
|
SIMDE_END_DECLS_
|
|
HEDLEY_DIAGNOSTIC_POP
|
|
|
|
#endif /* !defined(SIMDE_ARM_NEON_MMLAQ_H) */
|