arm_mat_trans_f32.c 7.0 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218
  1. /* ----------------------------------------------------------------------
  2. * Copyright (C) 2010-2014 ARM Limited. All rights reserved.
  3. *
  4. * $Date: 31. July 2014
  5. * $Revision: V1.4.4
  6. *
  7. * Project: CMSIS DSP Library
  8. * Title: arm_mat_trans_f32.c
  9. *
  10. * Description: Floating-point matrix transpose.
  11. *
  12. * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
  13. *
  14. * Redistribution and use in source and binary forms, with or without
  15. * modification, are permitted provided that the following conditions
  16. * are met:
  17. * - Redistributions of source code must retain the above copyright
  18. * notice, this list of conditions and the following disclaimer.
  19. * - Redistributions in binary form must reproduce the above copyright
  20. * notice, this list of conditions and the following disclaimer in
  21. * the documentation and/or other materials provided with the
  22. * distribution.
  23. * - Neither the name of ARM LIMITED nor the names of its contributors
  24. * may be used to endorse or promote products derived from this
  25. * software without specific prior written permission.
  26. *
  27. * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
  28. * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
  29. * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
  30. * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
  31. * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
  32. * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
  33. * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
  34. * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
  35. * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  36. * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
  37. * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
  38. * POSSIBILITY OF SUCH DAMAGE.
  39. * -------------------------------------------------------------------- */
  40. /**
  41. * @defgroup MatrixTrans Matrix Transpose
  42. *
  43. * Tranposes a matrix.
  44. * Transposing an <code>M x N</code> matrix flips it around the center diagonal and results in an <code>N x M</code> matrix.
  45. * \image html MatrixTranspose.gif "Transpose of a 3 x 3 matrix"
  46. */
  47. #include "arm_math.h"
  48. /**
  49. * @ingroup groupMatrix
  50. */
  51. /**
  52. * @addtogroup MatrixTrans
  53. * @{
  54. */
  55. /**
  56. * @brief Floating-point matrix transpose.
  57. * @param[in] *pSrc points to the input matrix
  58. * @param[out] *pDst points to the output matrix
  59. * @return The function returns either <code>ARM_MATH_SIZE_MISMATCH</code>
  60. * or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
  61. */
  62. arm_status arm_mat_trans_f32(
  63. const arm_matrix_instance_f32 * pSrc,
  64. arm_matrix_instance_f32 * pDst)
  65. {
  66. float32_t *pIn = pSrc->pData; /* input data matrix pointer */
  67. float32_t *pOut = pDst->pData; /* output data matrix pointer */
  68. float32_t *px; /* Temporary output data matrix pointer */
  69. uint16_t nRows = pSrc->numRows; /* number of rows */
  70. uint16_t nColumns = pSrc->numCols; /* number of columns */
  71. #ifndef ARM_MATH_CM0_FAMILY
  72. /* Run the below code for Cortex-M4 and Cortex-M3 */
  73. uint16_t blkCnt, i = 0u, row = nRows; /* loop counters */
  74. arm_status status; /* status of matrix transpose */
  75. #ifdef ARM_MATH_MATRIX_CHECK
  76. /* Check for matrix mismatch condition */
  77. if((pSrc->numRows != pDst->numCols) || (pSrc->numCols != pDst->numRows))
  78. {
  79. /* Set status as ARM_MATH_SIZE_MISMATCH */
  80. status = ARM_MATH_SIZE_MISMATCH;
  81. }
  82. else
  83. #endif /* #ifdef ARM_MATH_MATRIX_CHECK */
  84. {
  85. /* Matrix transpose by exchanging the rows with columns */
  86. /* row loop */
  87. do
  88. {
  89. /* Loop Unrolling */
  90. blkCnt = nColumns >> 2;
  91. /* The pointer px is set to starting address of the column being processed */
  92. px = pOut + i;
  93. /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
  94. ** a second loop below computes the remaining 1 to 3 samples. */
  95. while(blkCnt > 0u) /* column loop */
  96. {
  97. /* Read and store the input element in the destination */
  98. *px = *pIn++;
  99. /* Update the pointer px to point to the next row of the transposed matrix */
  100. px += nRows;
  101. /* Read and store the input element in the destination */
  102. *px = *pIn++;
  103. /* Update the pointer px to point to the next row of the transposed matrix */
  104. px += nRows;
  105. /* Read and store the input element in the destination */
  106. *px = *pIn++;
  107. /* Update the pointer px to point to the next row of the transposed matrix */
  108. px += nRows;
  109. /* Read and store the input element in the destination */
  110. *px = *pIn++;
  111. /* Update the pointer px to point to the next row of the transposed matrix */
  112. px += nRows;
  113. /* Decrement the column loop counter */
  114. blkCnt--;
  115. }
  116. /* Perform matrix transpose for last 3 samples here. */
  117. blkCnt = nColumns % 0x4u;
  118. while(blkCnt > 0u)
  119. {
  120. /* Read and store the input element in the destination */
  121. *px = *pIn++;
  122. /* Update the pointer px to point to the next row of the transposed matrix */
  123. px += nRows;
  124. /* Decrement the column loop counter */
  125. blkCnt--;
  126. }
  127. #else
  128. /* Run the below code for Cortex-M0 */
  129. uint16_t col, i = 0u, row = nRows; /* loop counters */
  130. arm_status status; /* status of matrix transpose */
  131. #ifdef ARM_MATH_MATRIX_CHECK
  132. /* Check for matrix mismatch condition */
  133. if((pSrc->numRows != pDst->numCols) || (pSrc->numCols != pDst->numRows))
  134. {
  135. /* Set status as ARM_MATH_SIZE_MISMATCH */
  136. status = ARM_MATH_SIZE_MISMATCH;
  137. }
  138. else
  139. #endif /* #ifdef ARM_MATH_MATRIX_CHECK */
  140. {
  141. /* Matrix transpose by exchanging the rows with columns */
  142. /* row loop */
  143. do
  144. {
  145. /* The pointer px is set to starting address of the column being processed */
  146. px = pOut + i;
  147. /* Initialize column loop counter */
  148. col = nColumns;
  149. while(col > 0u)
  150. {
  151. /* Read and store the input element in the destination */
  152. *px = *pIn++;
  153. /* Update the pointer px to point to the next row of the transposed matrix */
  154. px += nRows;
  155. /* Decrement the column loop counter */
  156. col--;
  157. }
  158. #endif /* #ifndef ARM_MATH_CM0_FAMILY */
  159. i++;
  160. /* Decrement the row loop counter */
  161. row--;
  162. } while(row > 0u); /* row loop end */
  163. /* Set status as ARM_MATH_SUCCESS */
  164. status = ARM_MATH_SUCCESS;
  165. }
  166. /* Return to application */
  167. return (status);
  168. }
  169. /**
  170. * @} end of MatrixTrans group
  171. */