/hal_nxp-3.5.0/mcux/mcux-sdk/CMSIS/DSP/Source/MatrixFunctions/ |
D | arm_mat_cmplx_mult_q31.c | 78 q63_t acc0, acc1, acc2, acc3; in arm_mat_cmplx_mult_q31_2x2_mve() local 97 acc3 = vmlaldavxq_s32(vecA, vecB); in arm_mat_cmplx_mult_q31_2x2_mve() 102 pOut[1 * CMPLX_DIM * MATRIX_DIM2 + 1] = (q31_t) asrl(acc3, 31); in arm_mat_cmplx_mult_q31_2x2_mve() 115 acc3 = vmlaldavxq_s32(vecA, vecB); in arm_mat_cmplx_mult_q31_2x2_mve() 122 pOut[1 * CMPLX_DIM * MATRIX_DIM2 + 1] = (q31_t) asrl(acc3, 31); in arm_mat_cmplx_mult_q31_2x2_mve() 141 q63_t acc0, acc1, acc2, acc3; in arm_mat_cmplx_mult_q31_3x3_mve() local 171 acc3 = vmlaldavxq_s32(vecA, vecB); in arm_mat_cmplx_mult_q31_3x3_mve() 179 acc3 = vmlaldavaxq_s32(acc3, vecA, vecB1); in arm_mat_cmplx_mult_q31_3x3_mve() 184 pOut[1 * CMPLX_DIM * MATRIX_DIM3 + 1] = (q31_t) asrl(acc3, 31); in arm_mat_cmplx_mult_q31_3x3_mve() 212 acc3 = vmlaldavxq_s32(vecA, vecB); in arm_mat_cmplx_mult_q31_3x3_mve() [all …]
|
D | arm_mat_cmplx_mult_f32.c | 305 f32x4_t acc0, acc1, acc2, acc3; in arm_mat_cmplx_mult_f32_4x4_mve() local 335 acc3 = vcmulq(vecA, vecB); in arm_mat_cmplx_mult_f32_4x4_mve() 336 acc3 = vcmlaq_rot90(acc3, vecA, vecB); in arm_mat_cmplx_mult_f32_4x4_mve() 353 acc3 = vcmlaq(acc3, vecA, vecB); in arm_mat_cmplx_mult_f32_4x4_mve() 354 acc3 = vcmlaq_rot90(acc3, vecA, vecB); in arm_mat_cmplx_mult_f32_4x4_mve() 362 pOut[3 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc3[0] + acc3[2]; in arm_mat_cmplx_mult_f32_4x4_mve() 363 pOut[3 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc3[1] + acc3[3]; in arm_mat_cmplx_mult_f32_4x4_mve() 386 acc3 = vcmulq(vecA, vecB); in arm_mat_cmplx_mult_f32_4x4_mve() 387 acc3 = vcmlaq_rot90(acc3, vecA, vecB); in arm_mat_cmplx_mult_f32_4x4_mve() 404 acc3 = vcmlaq(acc3, vecA, vecB); in arm_mat_cmplx_mult_f32_4x4_mve() [all …]
|
D | arm_mat_mult_q31.c | 225 q63_t acc0, acc1, acc2, acc3; in arm_mat_mult_q31_4x4_mve() local 241 acc3 = vrmlaldavhq(vecA, vecB); in arm_mat_mult_q31_4x4_mve() 246 acc3 = asrl(acc3, 23); in arm_mat_mult_q31_4x4_mve() 251 pOut[3 * MATRIX_DIM4] = (q31_t) acc3; in arm_mat_mult_q31_4x4_mve() 266 acc3 = vrmlaldavhq(vecA, vecB); in arm_mat_mult_q31_4x4_mve() 271 acc3 = asrl(acc3, 23); in arm_mat_mult_q31_4x4_mve() 276 pOut[3 * MATRIX_DIM4] = (q31_t) acc3; in arm_mat_mult_q31_4x4_mve() 292 acc3 = vrmlaldavhq(vecA, vecB); in arm_mat_mult_q31_4x4_mve() 297 acc3 = asrl(acc3, 23); in arm_mat_mult_q31_4x4_mve() 302 pOut[3 * MATRIX_DIM4] = (q31_t) acc3; in arm_mat_mult_q31_4x4_mve() [all …]
|
D | arm_mat_cmplx_mult_f16.c | 248 f16x8_t acc0, acc1, acc2, acc3; in arm_mat_cmplx_mult_f16_4x4_mve() local 275 acc3 = vcmulq(vecA, vecB); in arm_mat_cmplx_mult_f16_4x4_mve() 276 acc3 = vcmlaq_rot90(acc3, vecA, vecB); in arm_mat_cmplx_mult_f16_4x4_mve() 282 mve_cmplx_sum_intra_vec_f16(acc3, &pOut[3 * CMPLX_DIM * MATRIX_DIM]); in arm_mat_cmplx_mult_f16_4x4_mve() 304 acc3 = vcmulq(vecA, vecB); in arm_mat_cmplx_mult_f16_4x4_mve() 305 acc3 = vcmlaq_rot90(acc3, vecA, vecB); in arm_mat_cmplx_mult_f16_4x4_mve() 311 mve_cmplx_sum_intra_vec_f16(acc3, &pOut[3 * CMPLX_DIM * MATRIX_DIM]); in arm_mat_cmplx_mult_f16_4x4_mve() 333 acc3 = vcmulq(vecA, vecB); in arm_mat_cmplx_mult_f16_4x4_mve() 334 acc3 = vcmlaq_rot90(acc3, vecA, vecB); in arm_mat_cmplx_mult_f16_4x4_mve() 340 mve_cmplx_sum_intra_vec_f16(acc3, &pOut[3 * CMPLX_DIM * MATRIX_DIM]); in arm_mat_cmplx_mult_f16_4x4_mve() [all …]
|
D | arm_mat_mult_q15.c | 218 q63_t acc0, acc1, acc2, acc3; in arm_mat_mult_q15_4x4_mve() local 236 acc3 = vmlaldavq(vecA3, vecB); in arm_mat_mult_q15_4x4_mve() 241 acc3 = asrl(acc3, 15); in arm_mat_mult_q15_4x4_mve() 246 pOut[3 * MATRIX_DIM4] = (q15_t) __SSAT(acc3, 16); in arm_mat_mult_q15_4x4_mve() 257 acc3 = vmlaldavq(vecA3, vecB); in arm_mat_mult_q15_4x4_mve() 262 acc3 = asrl(acc3, 15); in arm_mat_mult_q15_4x4_mve() 267 pOut[3 * MATRIX_DIM4] = (q15_t) __SSAT(acc3, 16); in arm_mat_mult_q15_4x4_mve() 279 acc3 = vmlaldavq(vecA3, vecB); in arm_mat_mult_q15_4x4_mve() 284 acc3 = asrl(acc3, 15); in arm_mat_mult_q15_4x4_mve() 289 pOut[3 * MATRIX_DIM4] = (q15_t) __SSAT(acc3, 16); in arm_mat_mult_q15_4x4_mve() [all …]
|
/hal_nxp-3.5.0/mcux/mcux-sdk/CMSIS/DSP/Source/FilteringFunctions/ |
D | arm_conv_fast_q15.c | 73 q31_t sum, acc0, acc1, acc2, acc3; /* Accumulators */ in arm_conv_fast_q15() local 282 acc3 = 0; in arm_conv_fast_q15() 317 acc3 = __SMLADX(x3, c0, acc3); in arm_conv_fast_q15() 339 acc3 = __SMLADX(x1, c0, acc3); in arm_conv_fast_q15() 369 acc3 = __SMLADX(x3, c0, acc3); in arm_conv_fast_q15() 388 acc3 = __SMLADX(x2, c0, acc3); in arm_conv_fast_q15() 406 acc3 = __SMLADX(x2, c0, acc3); in arm_conv_fast_q15() 424 acc3 = __SMLADX(x3, c0, acc3); in arm_conv_fast_q15() 430 write_q15x2_ia (&pOut, __PKHBT((acc2 >> 15), (acc3 >> 15), 16)); in arm_conv_fast_q15() 433 write_q15x2_ia (&pOut, __PKHBT((acc3 >> 15), (acc2 >> 15), 16)); in arm_conv_fast_q15()
|
D | arm_correlate_fast_q15.c | 73 q31_t sum, acc0, acc1, acc2, acc3; /* Accumulators */ in arm_correlate_fast_q15() local 268 acc3 = 0; in arm_correlate_fast_q15() 303 acc3 = __SMLAD(x3, c0, acc3); in arm_correlate_fast_q15() 325 acc3 = __SMLAD(x1, c0, acc3); in arm_correlate_fast_q15() 355 acc3 = __SMLADX(x3, c0, acc3); in arm_correlate_fast_q15() 374 acc3 = __SMLAD(x2, c0, acc3); in arm_correlate_fast_q15() 392 acc3 = __SMLAD(x2, c0, acc3); in arm_correlate_fast_q15() 410 acc3 = __SMLADX(x3, c0, acc3); in arm_correlate_fast_q15() 424 *pOut = (q15_t) (acc3 >> 15); in arm_correlate_fast_q15()
|
D | arm_fir_decimate_fast_q31.c | 81 q63_t acc1, acc2, acc3; in arm_fir_decimate_fast_q31() local 109 acc3 = 0; in arm_fir_decimate_fast_q31() 141 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32); in arm_fir_decimate_fast_q31() 156 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32); in arm_fir_decimate_fast_q31() 171 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32); in arm_fir_decimate_fast_q31() 186 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32); in arm_fir_decimate_fast_q31() 210 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32); in arm_fir_decimate_fast_q31() 224 *pDst++ = (q31_t) (acc3 << 1); in arm_fir_decimate_fast_q31()
|
D | arm_conv_partial_fast_q15.c | 68 q31_t sum, acc0, acc1, acc2, acc3; /* Accumulator */ in arm_conv_partial_fast_q15() local 304 acc3 = 0; in arm_conv_partial_fast_q15() 341 acc3 = __SMLADX(x3, c0, acc3); in arm_conv_partial_fast_q15() 363 acc3 = __SMLADX(x1, c0, acc3); in arm_conv_partial_fast_q15() 392 acc3 = __SMLADX(x3, c0, acc3); in arm_conv_partial_fast_q15() 411 acc3 = __SMLADX(x2, c0, acc3); in arm_conv_partial_fast_q15() 429 acc3 = __SMLADX(x2, c0, acc3); in arm_conv_partial_fast_q15() 446 acc3 = __SMLADX(x3, c0, acc3); in arm_conv_partial_fast_q15() 452 write_q15x2_ia (&pOut, __PKHBT(acc2 >> 15, acc3 >> 15, 16)); in arm_conv_partial_fast_q15() 455 write_q15x2_ia (&pOut, __PKHBT(acc3 >> 15, acc2 >> 15, 16)); in arm_conv_partial_fast_q15()
|
D | arm_fir_fast_q15.c | 76 q31_t acc1, acc2, acc3; /* Accumulators */ in arm_fir_fast_q15() local 108 acc3 = 0; in arm_fir_fast_q15() 158 acc3 = __SMLADX(x1, c0, acc3); in arm_fir_fast_q15() 183 acc3 = __SMLADX(x1, c0, acc3); in arm_fir_fast_q15() 221 acc3 = __SMLADX(x1, c0, acc3); in arm_fir_fast_q15() 228 write_q15x2_ia (&pDst, __PKHBT(__SSAT((acc2 >> 15), 16), __SSAT((acc3 >> 15), 16), 16)); in arm_fir_fast_q15() 231 write_q15x2_ia (&pDst, __PKHBT(__SSAT((acc3 >> 15), 16), __SSAT((acc2 >> 15), 16), 16)); in arm_fir_fast_q15()
|
D | arm_conv_fast_q31.c | 76 q31_t sum, acc0, acc1, acc2, acc3; /* Accumulators */ in arm_conv_fast_q31() local 245 acc3 = 0; in arm_conv_fast_q31() 272 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32); in arm_conv_fast_q31() 288 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x0 * c0)) >> 32); in arm_conv_fast_q31() 304 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x1 * c0)) >> 32); in arm_conv_fast_q31() 320 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x2 * c0)) >> 32); in arm_conv_fast_q31() 344 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32); in arm_conv_fast_q31() 359 *pOut++ = (q31_t) (acc3 << 1); in arm_conv_fast_q31()
|
D | arm_fir_q31.c | 235 q63_t acc0=0, acc1=0, acc2=0, acc3=0; in arm_fir_q31_1_4_mve() local 263 acc3 = vrmlaldavhq(vecIn0, vecCoeffs); in arm_fir_q31_1_4_mve() 268 acc3 = asrl(acc3, 23); in arm_fir_q31_1_4_mve() 273 *pOutput++ = (q31_t) acc3; in arm_fir_q31_1_4_mve() 465 q63_t acc0, acc1, acc2, acc3; in arm_fir_q31_29_32_mve() local 561 acc3 = vrmlaldavhq(vecIn0, vecCoeffs0); in arm_fir_q31_29_32_mve() 563 acc3 = vrmlaldavhaq(acc3, vecIn0, vecCoeffs1); in arm_fir_q31_29_32_mve() 565 acc3 = vrmlaldavhaq(acc3, vecIn0, vecCoeffs2); in arm_fir_q31_29_32_mve() 567 acc3 = vrmlaldavhaq(acc3, vecIn0, vecCoeffs3); in arm_fir_q31_29_32_mve() 569 acc3 = vrmlaldavhaq(acc3, vecIn0, vecCoeffs4); in arm_fir_q31_29_32_mve() [all …]
|
D | arm_conv_partial_q15.c | 74 q63_t sum, acc0, acc1, acc2, acc3; /* Accumulator */ in arm_conv_partial_q15() local 310 acc3 = 0; in arm_conv_partial_q15() 347 acc3 = __SMLALDX(x3, c0, acc3); in arm_conv_partial_q15() 369 acc3 = __SMLALDX(x1, c0, acc3); in arm_conv_partial_q15() 398 acc3 = __SMLALDX(x3, c0, acc3); in arm_conv_partial_q15() 417 acc3 = __SMLALDX(x2, c0, acc3); in arm_conv_partial_q15() 435 acc3 = __SMLALDX(x2, c0, acc3); in arm_conv_partial_q15() 452 acc3 = __SMLALDX(x3, c0, acc3); in arm_conv_partial_q15() 458 write_q15x2_ia (&pOut, __PKHBT(__SSAT((acc2 >> 15), 16), __SSAT((acc3 >> 15), 16), 16)); in arm_conv_partial_q15() 461 write_q15x2_ia (&pOut, __PKHBT(__SSAT((acc3 >> 15), 16), __SSAT((acc2 >> 15), 16), 16)); in arm_conv_partial_q15()
|
D | arm_conv_q15.c | 146 int64_t acc3 = 0LL; in arm_conv_q15() local 155 MVE_INTR_CONV_QUAD_INC_X_FIXED_SIZE_Q15(acc0, acc1, acc2, acc3, pX, pY, count); in arm_conv_q15() 159 *pDst++ = (q15_t) acc3; in arm_conv_q15() 236 q63_t sum, acc0, acc1, acc2, acc3; /* Accumulators */ in arm_conv_q15() local 443 acc3 = 0; in arm_conv_q15() 479 acc3 = __SMLALDX(x3, c0, acc3); in arm_conv_q15() 502 acc3 = __SMLALDX(x1, c0, acc3); in arm_conv_q15() 531 acc3 = __SMLALDX(x3, c0, acc3); in arm_conv_q15() 550 acc3 = __SMLALDX(x2, c0, acc3); in arm_conv_q15() 568 acc3 = __SMLALDX(x2, c0, acc3); in arm_conv_q15() [all …]
|
D | arm_correlate_fast_q31.c | 78 q31_t sum, acc0, acc1, acc2, acc3; /* Accumulators */ in arm_correlate_fast_q31() local 271 acc3 = 0; in arm_correlate_fast_q31() 298 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32); in arm_correlate_fast_q31() 314 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x0 * c0)) >> 32); in arm_correlate_fast_q31() 330 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x1 * c0)) >> 32); in arm_correlate_fast_q31() 346 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x2 * c0)) >> 32); in arm_correlate_fast_q31() 370 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32); in arm_correlate_fast_q31() 392 *pOut = (q31_t) (acc3 << 1); in arm_correlate_fast_q31()
|
D | arm_conv_opt_q15.c | 82 q63_t acc1, acc2, acc3; /* Accumulators */ in arm_conv_opt_q15() local 202 acc3 = 0; in arm_conv_opt_q15() 247 acc3 = __SMLALDX(x3, y1, acc3); in arm_conv_opt_q15() 258 acc3 = __SMLALDX(x3, y2, acc3); in arm_conv_opt_q15() 276 acc3 += (*pScr1++ * *pIn2++); in arm_conv_opt_q15() 289 write_q15x2_ia (&pOut, __PKHBT(__SSAT((acc2 >> 15), 16), __SSAT((acc3 >> 15), 16), 16)); in arm_conv_opt_q15() 292 write_q15x2_ia (&pOut, __PKHBT(__SSAT((acc3 >> 15), 16), __SSAT((acc2 >> 15), 16), 16)); in arm_conv_opt_q15()
|
D | arm_conv_fast_opt_q15.c | 86 q31_t acc1, acc2, acc3; /* Accumulators */ in arm_conv_fast_opt_q15() local 206 acc3 = 0; in arm_conv_fast_opt_q15() 251 acc3 = __SMLADX(x3, y1, acc3); in arm_conv_fast_opt_q15() 262 acc3 = __SMLADX(x3, y2, acc3); in arm_conv_fast_opt_q15() 280 acc3 += (*pScr1++ * *pIn2++); in arm_conv_fast_opt_q15() 293 write_q15x2_ia (&pOut, __PKHBT(__SSAT((acc2 >> 15), 16), __SSAT((acc3 >> 15), 16), 16)); in arm_conv_fast_opt_q15() 296 write_q15x2_ia (&pOut, __PKHBT(__SSAT((acc3 >> 15), 16), __SSAT((acc2 >> 15), 16), 16)); in arm_conv_fast_opt_q15()
|
D | arm_conv_partial_fast_q31.c | 78 q31_t acc0, acc1, acc2, acc3; /* Accumulators */ in arm_conv_partial_fast_q31() local 280 acc3 = 0; in arm_conv_partial_fast_q31() 307 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32); in arm_conv_partial_fast_q31() 322 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x0 * c0)) >> 32); in arm_conv_partial_fast_q31() 337 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x1 * c0)) >> 32); in arm_conv_partial_fast_q31() 352 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x2 * c0)) >> 32); in arm_conv_partial_fast_q31() 375 acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32); in arm_conv_partial_fast_q31() 390 *pOut++ = (q31_t) (acc3 << 1); in arm_conv_partial_fast_q31()
|
D | arm_correlate_q15.c | 173 int64_t acc3 = 0LL; in arm_correlate_q15() local 182 MVE_INTR_CORR_QUAD_INC_X_FIXED_SIZE_Q15(acc0, acc1, acc2, acc3, pX, pY, srcBLen); in arm_correlate_q15() 190 *pDst = (q15_t) acc3; in arm_correlate_q15() 281 q63_t sum, acc0, acc1, acc2, acc3; /* Accumulators */ in arm_correlate_q15() local 474 acc3 = 0; in arm_correlate_q15() 510 acc3 = __SMLALD(x3, c0, acc3); in arm_correlate_q15() 532 acc3 = __SMLALD(x1, c0, acc3); in arm_correlate_q15() 558 acc3 = __SMLALDX(x3, c0, acc3); in arm_correlate_q15() 577 acc3 = __SMLALD(x2, c0, acc3); in arm_correlate_q15() 595 acc3 = __SMLALD(x2, c0, acc3); in arm_correlate_q15() [all …]
|
D | arm_conv_partial_fast_opt_q15.c | 85 q31_t acc1, acc2, acc3; /* Accumulator */ in arm_conv_partial_fast_opt_q15() local 216 acc3 = 0; in arm_conv_partial_fast_opt_q15() 261 acc3 = __SMLADX(x3, y1, acc3); in arm_conv_partial_fast_opt_q15() 273 acc3 = __SMLADX(x3, y2, acc3); in arm_conv_partial_fast_opt_q15() 291 acc3 += (*pScr1++ * *pIn2++); in arm_conv_partial_fast_opt_q15() 304 write_q15x2_ia (&pOut, __PKHBT(__SSAT((acc2 >> 15), 16), __SSAT((acc3 >> 15), 16), 16)); in arm_conv_partial_fast_opt_q15() 307 write_q15x2_ia (&pOut, __PKHBT(__SSAT((acc3 >> 15), 16), __SSAT((acc2 >> 15), 16), 16)); in arm_conv_partial_fast_opt_q15()
|
D | arm_conv_partial_opt_q15.c | 86 q63_t acc1, acc2, acc3; /* Accumulator */ in arm_conv_partial_opt_q15() local 217 acc3 = 0; in arm_conv_partial_opt_q15() 262 acc3 = __SMLALDX(x3, y1, acc3); in arm_conv_partial_opt_q15() 273 acc3 = __SMLALDX(x3, y2, acc3); in arm_conv_partial_opt_q15() 291 acc3 += (*pScr1++ * *pIn2++); in arm_conv_partial_opt_q15() 304 write_q15x2_ia (&pOut, __PKHBT(__SSAT((acc2 >> 15), 16), __SSAT((acc3 >> 15), 16), 16)); in arm_conv_partial_opt_q15() 307 write_q15x2_ia (&pOut, __PKHBT(__SSAT((acc3 >> 15), 16), __SSAT((acc2 >> 15), 16), 16)); in arm_conv_partial_opt_q15()
|
D | arm_fir_fast_q31.c | 78 q31_t acc1, acc2, acc3; /* Accumulators */ in arm_fir_fast_q31() local 110 acc3 = 0; in arm_fir_fast_q31() 147 multAcc_32x32_keep32_R(acc3, x3, c0); in arm_fir_fast_q31() 159 multAcc_32x32_keep32_R(acc3, x0, c0); in arm_fir_fast_q31() 171 multAcc_32x32_keep32_R(acc3, x1, c0); in arm_fir_fast_q31() 183 multAcc_32x32_keep32_R(acc3, x2, c0); in arm_fir_fast_q31() 208 multAcc_32x32_keep32_R(acc3, x3, c0); in arm_fir_fast_q31() 224 *pDst++ = (q31_t) (acc3 << 1); in arm_fir_fast_q31()
|
D | arm_correlate_fast_opt_q15.c | 83 q31_t acc1, acc2, acc3; /* Accumulators */ in arm_correlate_fast_opt_q15() local 184 acc3 = 0; in arm_correlate_fast_opt_q15() 228 acc3 = __SMLADX(x3, y1, acc3); in arm_correlate_fast_opt_q15() 239 acc3 = __SMLADX(x3, y2, acc3); in arm_correlate_fast_opt_q15() 257 acc3 += (*pScr1++ * *pIn2++); in arm_correlate_fast_opt_q15() 274 *pOut = (__SSAT(acc3 >> 15U, 16)); in arm_correlate_fast_opt_q15()
|
D | arm_correlate_opt_q15.c | 80 q63_t acc1, acc2, acc3; /* Accumulators */ in arm_correlate_opt_q15() local 179 acc3 = 0; in arm_correlate_opt_q15() 223 acc3 = __SMLALDX(x3, y1, acc3); in arm_correlate_opt_q15() 234 acc3 = __SMLALDX(x3, y2, acc3); in arm_correlate_opt_q15() 252 acc3 += (*pScr1++ * *pIn2++); in arm_correlate_opt_q15() 270 *pOut = (__SSAT(acc3 >> 15U, 16)); in arm_correlate_opt_q15()
|
D | arm_fir_q7.c | 207 q31_t acc0, acc1, acc2, acc3; in arm_fir_q7() local 264 acc3 = 0; in arm_fir_q7() 293 acc3 = vmladavaq(acc3, vecIn0, vecCoeffs); in arm_fir_q7() 308 *pOutput++ = (q7_t) __SSAT((acc3 >> 7U), 8); in arm_fir_q7() 463 q31_t acc1, acc2, acc3; /* Accumulators */ in arm_fir_q7() local 495 acc3 = 0; in arm_fir_q7() 532 acc3 += ((q15_t) x3 * c0); in arm_fir_q7() 544 acc3 += ((q15_t) x0 * c0); in arm_fir_q7() 556 acc3 += ((q15_t) x1 * c0); in arm_fir_q7() 568 acc3 += ((q15_t) x2 * c0); in arm_fir_q7() [all …]
|