| /hal_nxp-3.5.0/mcux/mcux-sdk/CMSIS/DSP/Source/BasicMathFunctions/ |
| D | arm_add_f32.c | 74 f32x4_t vec2; in arm_add_f32() local
86 vec2 = vld1q(pSrcB); in arm_add_f32()
87 res = vaddq(vec1, vec2); in arm_add_f32()
107 vec2 = vld1q(pSrcB); in arm_add_f32()
108 vstrwq_p(pDst, vaddq(vec1,vec2), p0); in arm_add_f32()
124 f32x4_t vec2; in arm_add_f32() local
136 vec2 = vld1q_f32(pSrcB); in arm_add_f32()
137 res = vaddq_f32(vec1, vec2); in arm_add_f32()
|
| D | arm_mult_f32.c | 74 f32x4_t vec2; in arm_mult_f32() local
85 vec2 = vld1q(pSrcB); in arm_mult_f32()
86 res = vmulq(vec1, vec2); in arm_mult_f32()
105 vec2 = vld1q(pSrcB); in arm_mult_f32()
106 vstrwq_p(pDst, vmulq(vec1,vec2), p0); in arm_mult_f32()
122 f32x4_t vec2; in arm_mult_f32() local
134 vec2 = vld1q_f32(pSrcB); in arm_mult_f32()
135 res = vmulq_f32(vec1, vec2); in arm_mult_f32()
|
| D | arm_sub_f32.c | 74 f32x4_t vec2; in arm_sub_f32() local
86 vec2 = vld1q(pSrcB); in arm_sub_f32()
87 res = vsubq(vec1, vec2); in arm_sub_f32()
107 vec2 = vld1q(pSrcB); in arm_sub_f32()
108 vstrwq_p(pDst, vsubq(vec1,vec2), p0); in arm_sub_f32()
124 f32x4_t vec2; in arm_sub_f32() local
136 vec2 = vld1q_f32(pSrcB); in arm_sub_f32()
137 res = vsubq_f32(vec1, vec2); in arm_sub_f32()
|
| D | arm_add_f16.c | 74 f16x8_t vec2; in arm_add_f16() local
86 vec2 = vld1q(pSrcB); in arm_add_f16()
87 res = vaddq(vec1, vec2); in arm_add_f16()
107 vec2 = vld1q(pSrcB); in arm_add_f16()
108 vstrhq_p(pDst, vaddq(vec1,vec2), p0); in arm_add_f16()
|
| D | arm_mult_f16.c | 75 f16x8_t vec2; in arm_mult_f16() local
86 vec2 = vld1q(pSrcB); in arm_mult_f16()
87 res = vmulq(vec1, vec2); in arm_mult_f16()
106 vec2 = vld1q(pSrcB); in arm_mult_f16()
107 vstrhq_p(pDst, vmulq(vec1,vec2), p0); in arm_mult_f16()
|
| D | arm_sub_f16.c | 73 f16x8_t vec2; in arm_sub_f16() local
85 vec2 = vld1q(pSrcB); in arm_sub_f16()
86 res = vsubq(vec1, vec2); in arm_sub_f16()
106 vec2 = vld1q(pSrcB); in arm_sub_f16()
107 vstrhq_p(pDst, vsubq(vec1,vec2), p0); in arm_sub_f16()
|
| D | arm_dot_prod_f32.c | 133 f32x4_t vec2; in arm_dot_prod_f32() local
141 vec2 = vld1q_f32(pSrcB); in arm_dot_prod_f32()
148 accum = vmlaq_f32(accum, vec1, vec2); in arm_dot_prod_f32()
155 vec2 = vld1q_f32(pSrcB); in arm_dot_prod_f32()
|
| /hal_nxp-3.5.0/mcux/mcux-sdk/CMSIS/DSP/Source/QuaternionMathFunctions/ |
| D | arm_quaternion2rotation_f32.c | 85 f32x4_t vec0,vec1, vec2 ,vec3; in arm_quaternion2rotation_f32() local
98 vec2 = vmulq_n_f32(vec0, vgetq_lane(vec0,0)); in arm_quaternion2rotation_f32()
101 vec2 = vmulq_n_f32(vec2, 2.0f); in arm_quaternion2rotation_f32()
117 vec0 = vsetq_lane(vgetq_lane(vec3,2) - vgetq_lane(vec2,3),vec0,1); in arm_quaternion2rotation_f32()
118 vec0 = vsetq_lane(vgetq_lane(vec3,3) + vgetq_lane(vec2,2),vec0,2); in arm_quaternion2rotation_f32()
119 vec0 = vsetq_lane(vgetq_lane(vec3,2) + vgetq_lane(vec2,3),vec0,3); in arm_quaternion2rotation_f32()
129 vec0 = vsetq_lane(q2q3 - vgetq_lane(vec2,1) ,vec0,1); in arm_quaternion2rotation_f32()
130 vec0 = vsetq_lane(vgetq_lane(vec3,3) - vgetq_lane(vec2,2),vec0,2); in arm_quaternion2rotation_f32()
131 vec0 = vsetq_lane(q2q3 + vgetq_lane(vec2,1) ,vec0,3); in arm_quaternion2rotation_f32()
|
| D | arm_quaternion_inverse_f32.c | 64 f32x4_t vec1,vec2; in arm_quaternion_inverse_f32() local
71 vec2 = vmulq(vec1,vec1); in arm_quaternion_inverse_f32()
72 squaredSum = vecAddAcrossF32Mve(vec2); in arm_quaternion_inverse_f32()
|
| D | arm_quaternion_normalize_f32.c | 63 f32x4_t vec1,vec2; in arm_quaternion_normalize_f32() local
69 vec2 = vmulq(vec1,vec1); in arm_quaternion_normalize_f32()
70 squaredSum = vecAddAcrossF32Mve(vec2); in arm_quaternion_normalize_f32()
|
| /hal_nxp-3.5.0/mcux/mcux-sdk/CMSIS/DSP/Source/MatrixFunctions/ |
| D | arm_mat_inverse_f16.c | 438 f16x8_t vec1, vec2; in arm_mat_inverse_f16() local
444 vec2 = vldrhq_f16(pPRT_in); in arm_mat_inverse_f16()
445 vec1 = vfmsq_f16(vec1, tmpV, vec2); in arm_mat_inverse_f16()
461 f16x8_t vec1, vec2; in arm_mat_inverse_f16() local
465 vec2 = vldrhq_f16(pPRT_in); in arm_mat_inverse_f16()
466 vec1 = vfmsq_f16(vec1, tmpV, vec2); in arm_mat_inverse_f16()
474 f16x8_t vec1, vec2; in arm_mat_inverse_f16() local
481 vec2 = vldrhq_f16(pPRT_pDst); in arm_mat_inverse_f16()
482 vec1 = vfmsq_f16(vec1, tmpV, vec2); in arm_mat_inverse_f16()
498 f16x8_t vec1, vec2; in arm_mat_inverse_f16() local
[all …]
|
| D | arm_mat_inverse_f32.c | 456 f32x4_t vec1, vec2; in arm_mat_inverse_f32() local
462 vec2 = vldrwq_f32(pPRT_in); in arm_mat_inverse_f32()
463 vec1 = vfmsq_f32(vec1, tmpV, vec2); in arm_mat_inverse_f32()
479 f32x4_t vec1, vec2; in arm_mat_inverse_f32() local
483 vec2 = vldrwq_f32(pPRT_in); in arm_mat_inverse_f32()
484 vec1 = vfmsq_f32(vec1, tmpV, vec2); in arm_mat_inverse_f32()
492 f32x4_t vec1, vec2; in arm_mat_inverse_f32() local
499 vec2 = vldrwq_f32(pPRT_pDst); in arm_mat_inverse_f32()
500 vec1 = vfmsq_f32(vec1, tmpV, vec2); in arm_mat_inverse_f32()
516 f32x4_t vec1, vec2; in arm_mat_inverse_f32() local
[all …]
|
| D | arm_mat_add_f32.c | 168 float32x4_t vec2; in arm_mat_add_f32() local
183 vec2 = vld1q_f32(pIn2); in arm_mat_add_f32()
184 res = vaddq_f32(vec1, vec2); in arm_mat_add_f32()
|
| D | arm_mat_sub_f32.c | 162 float32x4_t vec2; in arm_mat_sub_f32() local
178 vec2 = vld1q_f32(pIn2); in arm_mat_sub_f32()
179 res = vsubq_f32(vec1, vec2); in arm_mat_sub_f32()
|
| /hal_nxp-3.5.0/mcux/mcux-sdk/CMSIS/DSP/Source/ComplexMathFunctions/ |
| D | arm_cmplx_dot_prod_f32.c | 195 float32x4x2_t vec1,vec2,vec3,vec4; in arm_cmplx_dot_prod_f32() local
211 vec2 = vld2q_f32(pSrcB); in arm_cmplx_dot_prod_f32()
218 accR = vmlaq_f32(accR,vec1.val[0],vec2.val[0]); in arm_cmplx_dot_prod_f32()
219 accR = vmlsq_f32(accR,vec1.val[1],vec2.val[1]); in arm_cmplx_dot_prod_f32()
222 accI = vmlaq_f32(accI,vec1.val[1],vec2.val[0]); in arm_cmplx_dot_prod_f32()
223 accI = vmlaq_f32(accI,vec1.val[0],vec2.val[1]); in arm_cmplx_dot_prod_f32()
|
| /hal_nxp-3.5.0/mcux/mcux-sdk/CMSIS/DSP/Source/FilteringFunctions/ |
| D | arm_conv_f32.c | 340 float32x4_t vec2; in arm_conv_f32() local
368 vec2 = vld1q_f32(py-3); in arm_conv_f32()
369 vec2 = vrev64q_f32(vec2); in arm_conv_f32()
370 vec2 = vcombine_f32(vget_high_f32(vec2), vget_low_f32(vec2)); in arm_conv_f32()
372 res = vmlaq_f32(res,vec1, vec2); in arm_conv_f32()
|
| /hal_nxp-3.5.0/mcux/mcux-sdk/CMSIS/DSP/Source/SVMFunctions/ |
| D | arm_svm_linear_predict_f32.c | 300 float32x4_t vec2,vec2a,vec2b,vec2c,vec2d; in arm_svm_linear_predict_f32() local
408 vec2 = vld1q_f32(pSupport); in arm_svm_linear_predict_f32()
412 accum = vmlaq_f32(accum, vec1,vec2); in arm_svm_linear_predict_f32()
|
| D | arm_svm_sigmoid_predict_f32.c | 319 float32x4_t vec2,vec2a,vec2b,vec2c,vec2d; in arm_svm_sigmoid_predict_f32() local
433 vec2 = vld1q_f32(pSupport); in arm_svm_sigmoid_predict_f32()
437 accum = vmlaq_f32(accum, vec1,vec2); in arm_svm_sigmoid_predict_f32()
|
| D | arm_svm_polynomial_predict_f32.c | 321 float32x4_t vec2,vec2a,vec2b,vec2c,vec2d; in arm_svm_polynomial_predict_f32() local
435 vec2 = vld1q_f32(pSupport); in arm_svm_polynomial_predict_f32()
439 accum = vmlaq_f32(accum, vec1,vec2); in arm_svm_polynomial_predict_f32()
|
| D | arm_svm_rbf_predict_f32.c | 339 float32x4_t vec2,vec2a,vec2b,vec2c,vec2d; in arm_svm_rbf_predict_f32() local
464 vec2 = vld1q_f32(pSupport); in arm_svm_rbf_predict_f32()
468 temp = vsubq_f32(vec1,vec2); in arm_svm_rbf_predict_f32()
|
