| /cmsis-dsp-latest/Source/FilteringFunctions/ |
| D | arm_conv_fast_q31.c | 75 q31_t sum, acc0, acc1, acc2, acc3; /* Accumulators */ in arm_conv_fast_q31() local
149 sum = 0; in arm_conv_fast_q31()
159 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_fast_q31()
163 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_fast_q31()
167 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_fast_q31()
171 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_fast_q31()
185 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_fast_q31()
193 *pOut++ = sum << 1; in arm_conv_fast_q31()
378 sum = 0; in arm_conv_fast_q31()
388 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_fast_q31()
[all …]
|
| D | arm_correlate_fast_q31.c | 77 q31_t sum, acc0, acc1, acc2, acc3; /* Accumulators */ in arm_correlate_fast_q31() local
173 sum = 0; in arm_correlate_fast_q31()
183 sum = (q31_t) ((((q63_t) sum << 32) + in arm_correlate_fast_q31()
187 sum = (q31_t) ((((q63_t) sum << 32) + in arm_correlate_fast_q31()
191 sum = (q31_t) ((((q63_t) sum << 32) + in arm_correlate_fast_q31()
195 sum = (q31_t) ((((q63_t) sum << 32) + in arm_correlate_fast_q31()
210 sum = (q31_t) ((((q63_t) sum << 32) + in arm_correlate_fast_q31()
218 *pOut = sum << 1; in arm_correlate_fast_q31()
412 sum = 0; in arm_correlate_fast_q31()
422 sum = (q31_t) ((((q63_t) sum << 32) + in arm_correlate_fast_q31()
[all …]
|
| D | arm_conv_partial_fast_q31.c | 72 q31_t sum; /* Accumulators */ in arm_conv_partial_fast_q31() local
169 sum = 0; in arm_conv_partial_fast_q31()
179 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_partial_fast_q31()
183 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_partial_fast_q31()
187 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_partial_fast_q31()
191 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_partial_fast_q31()
211 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_partial_fast_q31()
219 *pOut++ = sum << 1; in arm_conv_partial_fast_q31()
416 sum = 0; in arm_conv_partial_fast_q31()
426 sum = (q31_t) ((((q63_t) sum << 32) + in arm_conv_partial_fast_q31()
[all …]
|
| D | arm_conv_fast_q15.c | 72 q31_t sum, acc0, acc1, acc2, acc3; /* Accumulators */ in arm_conv_fast_q15() local
153 sum = 0; in arm_conv_fast_q15()
162 sum = __SMLAD(*px++, *py--, sum); in arm_conv_fast_q15()
169 *pOut++ = (q15_t) (sum >> 15); in arm_conv_fast_q15()
191 sum = 0; in arm_conv_fast_q15()
202 sum = __SMLADX(read_q15x2_ia ((q15_t **) &px), read_q15x2_da ((q15_t **) &py), sum); in arm_conv_fast_q15()
204 sum = __SMLADX(read_q15x2_ia ((q15_t **) &px), read_q15x2_da ((q15_t **) &py), sum); in arm_conv_fast_q15()
221 sum = __SMLAD(*px++, *py--, sum); in arm_conv_fast_q15()
228 *pOut++ = (q15_t) (sum >> 15); in arm_conv_fast_q15()
453 sum = 0; in arm_conv_fast_q15()
[all …]
|
| D | arm_conv_partial_q15.c | 74 q63_t sum, acc0, acc1, acc2, acc3; /* Accumulator */ in arm_conv_partial_q15() local
174 sum = 0; in arm_conv_partial_q15()
183 sum = __SMLALD(*px++, *py--, sum); in arm_conv_partial_q15()
190 *pOut++ = (q15_t) (__SSAT((sum >> 15), 16)); in arm_conv_partial_q15()
212 sum = 0; in arm_conv_partial_q15()
223 sum = __SMLALDX(read_q15x2_ia ((q15_t **) &px), read_q15x2_da ((q15_t **) &py), sum); in arm_conv_partial_q15()
225 sum = __SMLALDX(read_q15x2_ia ((q15_t **) &px), read_q15x2_da ((q15_t **) &py), sum); in arm_conv_partial_q15()
242 sum = __SMLALD(*px++, *py--, sum); in arm_conv_partial_q15()
249 *pOut++ = (q15_t) (__SSAT((sum >> 15), 16)); in arm_conv_partial_q15()
488 sum = 0; in arm_conv_partial_q15()
[all …]
|
| D | arm_conv_partial_fast_q15.c | 68 q31_t sum, acc0, acc1, acc2, acc3; /* Accumulator */ in arm_conv_partial_fast_q15() local
168 sum = 0; in arm_conv_partial_fast_q15()
177 sum = __SMLAD(*px++, *py--, sum); in arm_conv_partial_fast_q15()
184 *pOut++ = (q15_t) (sum >> 15); in arm_conv_partial_fast_q15()
206 sum = 0; in arm_conv_partial_fast_q15()
217 sum = __SMLADX(read_q15x2_ia ((q15_t **) &px), read_q15x2_da ((q15_t **) &py), sum); in arm_conv_partial_fast_q15()
219 sum = __SMLADX(read_q15x2_ia ((q15_t **) &px), read_q15x2_da ((q15_t **) &py), sum); in arm_conv_partial_fast_q15()
236 sum = __SMLAD(*px++, *py--, sum); in arm_conv_partial_fast_q15()
243 *pOut++ = (q15_t) (sum >> 15); in arm_conv_partial_fast_q15()
476 sum = 0; in arm_conv_partial_fast_q15()
[all …]
|
| D | arm_conv_partial_f32.c | 102 float32_t sum; /* Accumulator */ in arm_conv_partial_f32() local
199 sum = 0.0f; in arm_conv_partial_f32()
209 sum += *px++ * *py--; in arm_conv_partial_f32()
212 sum += *px++ * *py--; in arm_conv_partial_f32()
215 sum += *px++ * *py--; in arm_conv_partial_f32()
218 sum += *px++ * *py--; in arm_conv_partial_f32()
237 sum += *px++ * *py--; in arm_conv_partial_f32()
244 *pOut++ = sum; in arm_conv_partial_f32()
441 sum = 0.0f; in arm_conv_partial_f32()
451 sum += *px++ * *py--; in arm_conv_partial_f32()
[all …]
|
| D | arm_conv_partial_q31.c | 75 q63_t sum; /* Accumulator */ in arm_conv_partial_q31() local
172 sum = 0; in arm_conv_partial_q31()
182 sum += (q63_t) *px++ * (*py--); in arm_conv_partial_q31()
185 sum += (q63_t) *px++ * (*py--); in arm_conv_partial_q31()
188 sum += (q63_t) *px++ * (*py--); in arm_conv_partial_q31()
191 sum += (q63_t) *px++ * (*py--); in arm_conv_partial_q31()
210 sum += (q63_t) *px++ * (*py--); in arm_conv_partial_q31()
217 *pOut++ = (q31_t) (sum >> 31); in arm_conv_partial_q31()
395 sum = 0; in arm_conv_partial_q31()
405 sum += (q63_t) *px++ * (*py--); in arm_conv_partial_q31()
[all …]
|
| D | arm_correlate_fast_q15.c | 72 q31_t sum, acc0, acc1, acc2, acc3; /* Accumulators */ in arm_correlate_fast_q15() local
182 sum = 0; in arm_correlate_fast_q15()
192 sum = __SMLAD(read_q15x2_ia ((q15_t **) &px), read_q15x2_ia ((q15_t **) &py), sum); in arm_correlate_fast_q15()
194 sum = __SMLAD(read_q15x2_ia ((q15_t **) &px), read_q15x2_ia ((q15_t **) &py), sum); in arm_correlate_fast_q15()
208 sum = __SMLAD(*px++, *py++, sum); in arm_correlate_fast_q15()
215 *pOut = (q15_t) (sum >> 15); in arm_correlate_fast_q15()
444 sum = 0; in arm_correlate_fast_q15()
454 sum += ((q31_t) *px++ * *py++); in arm_correlate_fast_q15()
455 sum += ((q31_t) *px++ * *py++); in arm_correlate_fast_q15()
456 sum += ((q31_t) *px++ * *py++); in arm_correlate_fast_q15()
[all …]
|
| D | arm_conv_q15.c | 242 q63_t sum, acc0, acc1, acc2, acc3; /* Accumulators */ in arm_conv_q15() local
326 sum = 0; in arm_conv_q15()
335 sum = __SMLALD(*px++, *py--, sum); in arm_conv_q15()
342 *pOut++ = (q15_t) (__SSAT((sum >> 15), 16)); in arm_conv_q15()
364 sum = 0; in arm_conv_q15()
375 sum = __SMLALDX(read_q15x2_ia ((q15_t **) &px), read_q15x2_da ((q15_t **) &py), sum); in arm_conv_q15()
377 sum = __SMLALDX(read_q15x2_ia ((q15_t **) &px), read_q15x2_da ((q15_t **) &py), sum); in arm_conv_q15()
394 sum = __SMLALD(*px++, *py--, sum); in arm_conv_q15()
401 *pOut++ = (q15_t) (__SSAT((sum >> 15), 16)); in arm_conv_q15()
631 sum = 0; in arm_conv_q15()
[all …]
|
| /cmsis-dsp-latest/Source/BasicMathFunctions/ |
| D | arm_dot_prod_q15.c | 67 q63_t sum = 0LL; in arm_dot_prod_q15() local
79 sum = vmlaldavaq(sum, vecA, vecB); in arm_dot_prod_q15()
99 sum = vmlaldavaq_p(sum, vecA, vecB, p0); in arm_dot_prod_q15()
102 *result = sum; in arm_dot_prod_q15()
113 q63_t sum = 0; /* Temporary return variable */ in arm_dot_prod_q15() local
126 sum = __SMLALD(read_q15x2_ia (&pSrcA), read_q15x2_ia (&pSrcB), sum); in arm_dot_prod_q15()
127 sum = __SMLALD(read_q15x2_ia (&pSrcA), read_q15x2_ia (&pSrcB), sum); in arm_dot_prod_q15()
129 sum += (q63_t)((q31_t) *pSrcA++ * *pSrcB++); in arm_dot_prod_q15()
130 sum += (q63_t)((q31_t) *pSrcA++ * *pSrcB++); in arm_dot_prod_q15()
131 sum += (q63_t)((q31_t) *pSrcA++ * *pSrcB++); in arm_dot_prod_q15()
[all …]
|
| D | arm_dot_prod_q7.c | 68 q31_t sum = 0; in arm_dot_prod_q7() local
80 sum = vmladavaq(sum, vecA, vecB); in arm_dot_prod_q7()
100 sum = vmladavaq_p(sum, vecA, vecB, p0); in arm_dot_prod_q7()
103 *result = sum; in arm_dot_prod_q7()
113 q31_t sum = 0; /* Temporary return variable */ in arm_dot_prod_q7() local
145 sum = __SMLAD(inA1, inB1, sum); in arm_dot_prod_q7()
146 sum = __SMLAD(inA2, inB2, sum); in arm_dot_prod_q7()
148 sum += (q31_t) ((q15_t) *pSrcA++ * *pSrcB++); in arm_dot_prod_q7()
149 sum += (q31_t) ((q15_t) *pSrcA++ * *pSrcB++); in arm_dot_prod_q7()
150 sum += (q31_t) ((q15_t) *pSrcA++ * *pSrcB++); in arm_dot_prod_q7()
[all …]
|
| D | arm_dot_prod_q31.c | 69 q63_t sum = 0LL; in arm_dot_prod_q31() local
81 sum = vrmlaldavhaq(sum, vecA, vecB); in arm_dot_prod_q31()
101 sum = vrmlaldavhaq_p(sum, vecA, vecB, p0); in arm_dot_prod_q31()
110 *result = asrl(sum, (14 - 8)); in arm_dot_prod_q31()
121 q63_t sum = 0; /* Temporary return variable */ in arm_dot_prod_q31() local
133 sum += ((q63_t) *pSrcA++ * *pSrcB++) >> 14U; in arm_dot_prod_q31()
135 sum += ((q63_t) *pSrcA++ * *pSrcB++) >> 14U; in arm_dot_prod_q31()
137 sum += ((q63_t) *pSrcA++ * *pSrcB++) >> 14U; in arm_dot_prod_q31()
139 sum += ((q63_t) *pSrcA++ * *pSrcB++) >> 14U; in arm_dot_prod_q31()
160 sum += ((q63_t) *pSrcA++ * *pSrcB++) >> 14U; in arm_dot_prod_q31()
[all …]
|
| /cmsis-dsp-latest/Source/StatisticsFunctions/ |
| D | arm_power_q15.c | 64 q63_t sum = 0LL; in arm_power_q15() local
75 sum = vmlaldavaq(sum, vecSrc, vecSrc); in arm_power_q15()
91 sum += ((q31_t) in * in); in arm_power_q15()
97 *pResult = sum; in arm_power_q15()
106 q63_t sum = 0; /* Temporary result storage */ in arm_power_q15() local
125 sum = __SMLALD(in32, in32, sum); in arm_power_q15()
128 sum = __SMLALD(in32, in32, sum); in arm_power_q15()
131 sum += ((q31_t) in * in); in arm_power_q15()
134 sum += ((q31_t) in * in); in arm_power_q15()
137 sum += ((q31_t) in * in); in arm_power_q15()
[all …]
|
| D | arm_power_q7.c | 63 q31_t sum = 0LL; in arm_power_q7() local
74 sum = vmladavaq(sum, vecSrc, vecSrc); in arm_power_q7()
90 sum += ((q15_t) in * in); in arm_power_q7()
96 *pResult = sum; in arm_power_q7()
105 q31_t sum = 0; /* Temporary result storage */ in arm_power_q7() local
130 sum = __SMLAD(in1, in1, sum); in arm_power_q7()
131 sum = __SMLAD(in2, in2, sum); in arm_power_q7()
134 sum += ((q15_t) in * in); in arm_power_q7()
137 sum += ((q15_t) in * in); in arm_power_q7()
140 sum += ((q15_t) in * in); in arm_power_q7()
[all …]
|
| D | arm_var_q15.c | 66 q63_t sum = 0LL; in arm_var_q15() local
84 sum = vaddvaq_s16(sum, vecSrc); in arm_var_q15()
105 sum += in; in arm_var_q15()
116 squareOfMean = arm_div_int64_to_int32((q63_t)sum * sum, (q31_t)(blockSize * (blockSize - 1U))); in arm_var_q15()
128 q31_t sum = 0; /* Accumulator */ in arm_var_q15() local
158 sum += ((in32 << 16U) >> 16U); in arm_var_q15()
159 sum += (in32 >> 16U); in arm_var_q15()
163 sum += ((in32 << 16U) >> 16U); in arm_var_q15()
164 sum += (in32 >> 16U); in arm_var_q15()
168 sum += in; in arm_var_q15()
[all …]
|
| D | arm_mean_f32.c | 59 float32_t sum = 0.0f; in arm_mean_f32() local
72 sum = vecAddAcrossF32Mve(sumVec); in arm_mean_f32()
80 sum += *pSrc++; in arm_mean_f32()
86 *pResult = sum / (float32_t) blockSize; in arm_mean_f32()
97 float32_t sum = 0.0f; /* Temporary result storage */ in arm_mean_f32() local
121 sum = vget_lane_f32(sumV2, 0) + vget_lane_f32(sumV2, 1); in arm_mean_f32()
130 sum += *pSrc++; in arm_mean_f32()
138 *pResult = sum / (float32_t) blockSize; in arm_mean_f32()
147 float32_t sum = 0.0f; /* Temporary result storage */ in arm_mean_f32() local
157 sum += *pSrc++; in arm_mean_f32()
[all …]
|
| D | arm_power_f32.c | 75 float32_t sum = 0.0f; in arm_power_f32() local
91 sum = vecAddAcrossF32Mve(sumVec); in arm_power_f32()
103 sum += in * in; in arm_power_f32()
109 *pResult = sum; in arm_power_f32()
118 float32_t sum = 0.0f; /* accumulator */ in arm_power_f32() local
142 sum = vget_lane_f32(sumV2, 0) + vget_lane_f32(sumV2, 1); in arm_power_f32()
153 sum += in * in; in arm_power_f32()
160 *pResult = sum; in arm_power_f32()
169 float32_t sum = 0.0f; /* Temporary result storage */ in arm_power_f32() local
183 sum += in * in; in arm_power_f32()
[all …]
|
| D | arm_mean_q31.c | 62 q63_t sum = 0LL; in arm_mean_q31() local
74 sum = vaddlvaq(sum, vecSrc); in arm_mean_q31()
86 sum += *pSrc++; in arm_mean_q31()
90 *pResult = arm_div_int64_to_int32(sum, blockSize); in arm_mean_q31()
99 q63_t sum = 0; /* Temporary result storage */ in arm_mean_q31() local
109 sum += *pSrc++; in arm_mean_q31()
111 sum += *pSrc++; in arm_mean_q31()
113 sum += *pSrc++; in arm_mean_q31()
115 sum += *pSrc++; in arm_mean_q31()
134 sum += *pSrc++; in arm_mean_q31()
[all …]
|
| D | arm_mean_q15.c | 63 q31_t sum = 0L; in arm_mean_q15() local
73 sum = vaddvaq(sum, vecSrc); in arm_mean_q15()
85 sum += *pSrc++; in arm_mean_q15()
93 *pResult = (q15_t) (sum / (int32_t) blockSize); in arm_mean_q15()
102 q31_t sum = 0; /* Temporary result storage */ in arm_mean_q15() local
117 sum += ((in << 16U) >> 16U); in arm_mean_q15()
118 sum += (in >> 16U); in arm_mean_q15()
121 sum += ((in << 16U) >> 16U); in arm_mean_q15()
122 sum += (in >> 16U); in arm_mean_q15()
141 sum += *pSrc++; in arm_mean_q15()
[all …]
|
| D | arm_mean_q7.c | 64 q31_t sum = 0L; in arm_mean_q7() local
74 sum = vaddvaq(sum, vecSrc); in arm_mean_q7()
84 sum += *pSrc++; in arm_mean_q7()
92 *pResult = (q7_t) (sum / (int32_t) blockSize); in arm_mean_q7()
101 q31_t sum = 0; /* Temporary result storage */ in arm_mean_q7() local
116 sum += ((in << 24U) >> 24U); in arm_mean_q7()
117 sum += ((in << 16U) >> 24U); in arm_mean_q7()
118 sum += ((in << 8U) >> 24U); in arm_mean_q7()
119 sum += (in >> 24U); in arm_mean_q7()
138 sum += *pSrc++; in arm_mean_q7()
[all …]
|
| D | arm_accumulate_f32.c | 60 float32_t sum = 0.0f; in arm_accumulate_f32() local
93 sum = vecAddAcrossF32Mve(vecSum); in arm_accumulate_f32()
96 *pResult = sum; in arm_accumulate_f32()
107 float32_t sum = 0.0f; /* Temporary result storage */ in arm_accumulate_f32() local
131 sum = vget_lane_f32(sumV2, 0) + vget_lane_f32(sumV2, 1); in arm_accumulate_f32()
140 sum += *pSrc++; in arm_accumulate_f32()
148 *pResult = sum; in arm_accumulate_f32()
158 float32_t sum = 0.0f; /* Temporary result storage */ in arm_accumulate_f32() local
168 sum += *pSrc++; in arm_accumulate_f32()
170 sum += *pSrc++; in arm_accumulate_f32()
[all …]
|
| D | arm_var_q31.c | 69 q63_t sum = 0LL; in arm_var_q31() local
90 sum = vaddlvaq(sum, vecSrc); in arm_var_q31()
110 sum += in; in arm_var_q31()
121 squareOfMean = sum * sum / (q63_t) (blockSize * (blockSize - 1U)); in arm_var_q31()
133 q63_t sum = 0; /* Temporary result storage */ in arm_var_q31() local
158 sum += in; in arm_var_q31()
162 sum += in; in arm_var_q31()
166 sum += in; in arm_var_q31()
170 sum += in; in arm_var_q31()
195 sum += in; in arm_var_q31()
[all …]
|
| /cmsis-dsp-latest/Source/ComplexMathFunctions/ |
| D | arm_cmplx_mag_f16.c | 63 f16x8_t sum; in arm_cmplx_mag_f16() local
74 sum = vmulq(vecSrc.val[0], vecSrc.val[0]); in arm_cmplx_mag_f16()
75 sum = vfmaq(sum, vecSrc.val[1], vecSrc.val[1]); in arm_cmplx_mag_f16()
82 newtonStartVec = vdupq_n_s16(INVSQRT_MAGIC_F16) - vshrq((q15x8_t) sum, 1); in arm_cmplx_mag_f16()
83 sumHalf = sum * 0.5f; in arm_cmplx_mag_f16()
101 sum = vmulq(sum, invSqrt); in arm_cmplx_mag_f16()
102 vstrhq_f16(pDst, sum); in arm_cmplx_mag_f16()
120 sum = vmulq(vecSrc.val[0], vecSrc.val[0]); in arm_cmplx_mag_f16()
121 sum = vfmaq(sum, vecSrc.val[1], vecSrc.val[1]); in arm_cmplx_mag_f16()
128 newtonStartVec = vdupq_n_s16(INVSQRT_MAGIC_F16) - vshrq((q15x8_t) sum, 1); in arm_cmplx_mag_f16()
[all …]
|
| /cmsis-dsp-latest/Source/MatrixFunctions/ |
| D | arm_mat_mult_fast_q15.c | 73 q31_t sum; /* Accumulator */ in arm_mat_mult_fast_q15() local
248 sum = 0; in arm_mat_mult_fast_q15()
281 sum = __SMLAD(inA1, inB1, sum); in arm_mat_mult_fast_q15()
290 sum += inA1 * inB1; in arm_mat_mult_fast_q15()
294 sum += inA2 * inB2; in arm_mat_mult_fast_q15()
298 sum += inA1 * inB1; in arm_mat_mult_fast_q15()
302 sum += inA2 * inB2; in arm_mat_mult_fast_q15()
316 sum += inA1 * inB1; in arm_mat_mult_fast_q15()
327 sum += (q31_t) *pInA++ * *pInB++; in arm_mat_mult_fast_q15()
335 *px++ = (q15_t) (sum >> 15); in arm_mat_mult_fast_q15()
[all …]
|
