arm_abs_f32.c 3.4 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146
  1. /* ----------------------------------------------------------------------
  2. * Project: CMSIS DSP Library
  3. * Title: arm_abs_f32.c
  4. * Description: Floating-point vector absolute value
  5. *
  6. * $Date: 18. March 2019
  7. * $Revision: V1.6.0
  8. *
  9. * Target Processor: Cortex-M cores
  10. * -------------------------------------------------------------------- */
  11. /*
  12. * Copyright (C) 2010-2019 ARM Limited or its affiliates. All rights reserved.
  13. *
  14. * SPDX-License-Identifier: Apache-2.0
  15. *
  16. * Licensed under the Apache License, Version 2.0 (the License); you may
  17. * not use this file except in compliance with the License.
  18. * You may obtain a copy of the License at
  19. *
  20. * www.apache.org/licenses/LICENSE-2.0
  21. *
  22. * Unless required by applicable law or agreed to in writing, software
  23. * distributed under the License is distributed on an AS IS BASIS, WITHOUT
  24. * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  25. * See the License for the specific language governing permissions and
  26. * limitations under the License.
  27. */
  28. #include "arm_math.h"
  29. #include <math.h>
  30. /**
  31. @ingroup groupMath
  32. */
  33. /**
  34. @defgroup BasicAbs Vector Absolute Value
  35. Computes the absolute value of a vector on an element-by-element basis.
  36. <pre>
  37. pDst[n] = abs(pSrc[n]), 0 <= n < blockSize.
  38. </pre>
  39. The functions support in-place computation allowing the source and
  40. destination pointers to reference the same memory buffer.
  41. There are separate functions for floating-point, Q7, Q15, and Q31 data types.
  42. */
  43. /**
  44. @addtogroup BasicAbs
  45. @{
  46. */
  47. /**
  48. @brief Floating-point vector absolute value.
  49. @param[in] pSrc points to the input vector
  50. @param[out] pDst points to the output vector
  51. @param[in] blockSize number of samples in each vector
  52. @return none
  53. */
  54. void arm_abs_f32(
  55. const float32_t * pSrc,
  56. float32_t * pDst,
  57. uint32_t blockSize)
  58. {
  59. uint32_t blkCnt; /* Loop counter */
  60. #if defined(ARM_MATH_NEON)
  61. float32x4_t vec1;
  62. float32x4_t res;
  63. /* Compute 4 outputs at a time */
  64. blkCnt = blockSize >> 2U;
  65. while (blkCnt > 0U)
  66. {
  67. /* C = |A| */
  68. /* Calculate absolute values and then store the results in the destination buffer. */
  69. vec1 = vld1q_f32(pSrc);
  70. res = vabsq_f32(vec1);
  71. vst1q_f32(pDst, res);
  72. /* Increment pointers */
  73. pSrc += 4;
  74. pDst += 4;
  75. /* Decrement the loop counter */
  76. blkCnt--;
  77. }
  78. /* Tail */
  79. blkCnt = blockSize & 0x3;
  80. #else
  81. #if defined (ARM_MATH_LOOPUNROLL)
  82. /* Loop unrolling: Compute 4 outputs at a time */
  83. blkCnt = blockSize >> 2U;
  84. while (blkCnt > 0U)
  85. {
  86. /* C = |A| */
  87. /* Calculate absolute and store result in destination buffer. */
  88. *pDst++ = fabsf(*pSrc++);
  89. *pDst++ = fabsf(*pSrc++);
  90. *pDst++ = fabsf(*pSrc++);
  91. *pDst++ = fabsf(*pSrc++);
  92. /* Decrement loop counter */
  93. blkCnt--;
  94. }
  95. /* Loop unrolling: Compute remaining outputs */
  96. blkCnt = blockSize % 0x4U;
  97. #else
  98. /* Initialize blkCnt with number of samples */
  99. blkCnt = blockSize;
  100. #endif /* #if defined (ARM_MATH_LOOPUNROLL) */
  101. #endif /* #if defined(ARM_MATH_NEON) */
  102. while (blkCnt > 0U)
  103. {
  104. /* C = |A| */
  105. /* Calculate absolute and store result in destination buffer. */
  106. *pDst++ = fabsf(*pSrc++);
  107. /* Decrement loop counter */
  108. blkCnt--;
  109. }
  110. }
  111. /**
  112. @} end of BasicAbs group
  113. */