arm_mat_trans_f32.c

00001 /* ----------------------------------------------------------------------  
00002 * Copyright (C) 2010 ARM Limited. All rights reserved.  
00003 *  
00004 * $Date:        29. November 2010  
00005 * $Revision:    V1.0.3  
00006 *  
00007 * Project:      CMSIS DSP Library  
00008 * Title:        arm_mat_trans_f32.c  
00009 *  
00010 * Description:  Floating-point matrix transpose.  
00011 *  
00012 * Target Processor: Cortex-M4/Cortex-M3
00013 *  
00014 * Version 1.0.3 2010/11/29 
00015 *    Re-organized the CMSIS folders and updated documentation.  
00016 *   
00017 * Version 1.0.2 2010/11/11  
00018 *    Documentation updated.   
00019 *  
00020 * Version 1.0.1 2010/10/05   
00021 *    Production release and review comments incorporated.  
00022 *  
00023 * Version 1.0.0 2010/09/20   
00024 *    Production release and review comments incorporated.  
00025 *  
00026 * Version 0.0.5  2010/04/26   
00027 *    incorporated review comments and updated with latest CMSIS layer  
00028 *  
00029 * Version 0.0.3  2010/03/10   
00030 *    Initial version  
00031 * -------------------------------------------------------------------- */ 
00032  
00033 /**  
00034  * @defgroup MatrixTrans Matrix Transpose  
00035  *  
00036  * Tranposes a matrix.  
00037  * Transposing an <code>M x N</code> matrix flips it around the center diagonal and results in an <code>N x M</code> matrix.  
00038  * \image html MatrixTranspose.gif "Transpose of a 3 x 3 matrix"  
00039  */ 
00040  
00041 #include "arm_math.h" 
00042  
00043 /**  
00044  * @ingroup groupMatrix  
00045  */ 
00046  
00047 /**  
00048  * @addtogroup MatrixTrans  
00049  * @{  
00050  */ 
00051  
00052 /**  
00053   * @brief Floating-point matrix transpose.  
00054   * @param[in]  *pSrc points to the input matrix  
00055   * @param[out] *pDst points to the output matrix  
00056   * @return     The function returns either  <code>ARM_MATH_SIZE_MISMATCH</code>  
00057   * or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.  
00058   */ 
00059  
00060  
00061 arm_status arm_mat_trans_f32( 
00062   const arm_matrix_instance_f32 * pSrc, 
00063   arm_matrix_instance_f32 * pDst) 
00064 { 
00065   float32_t *pIn = pSrc->pData;                  /* input data matrix pointer */ 
00066   float32_t *pOut = pDst->pData;                 /* output data matrix pointer */ 
00067   float32_t *px;                                 /* Temporary output data matrix pointer */ 
00068   uint16_t nRows = pSrc->numRows;                /* number of rows */ 
00069   uint16_t nColumns = pSrc->numCols;             /* number of columns */ 
00070   uint16_t blkCnt, i = 0u, row = nRows;          /* loop counters */ 
00071   arm_status status;                             /* status of matrix transpose  */ 
00072  
00073  
00074 #ifdef ARM_MATH_MATRIX_CHECK 
00075   /* Check for matrix mismatch condition */ 
00076   if((pSrc->numRows != pDst->numCols) || (pSrc->numCols != pDst->numRows)) 
00077   { 
00078     /* Set status as ARM_MATH_SIZE_MISMATCH */ 
00079     status = ARM_MATH_SIZE_MISMATCH; 
00080   } 
00081   else 
00082 #endif 
00083   { 
00084     /* Matrix transpose by exchanging the rows with columns */ 
00085     /* row loop     */ 
00086     do 
00087     { 
00088       /* Loop Unrolling */ 
00089       blkCnt = nColumns >> 2; 
00090  
00091       /* The pointer px is set to starting address of the column being processed */ 
00092       px = pOut + i; 
00093  
00094       /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00095        ** a second loop below computes the remaining 1 to 3 samples. */ 
00096       while(blkCnt > 0u)        /* column loop */ 
00097       { 
00098         /* Read and store the input element in the destination */ 
00099         *px = *pIn++; 
00100  
00101         /* Update the pointer px to point to the next row of the transposed matrix */ 
00102         px += nRows; 
00103  
00104         /* Read and store the input element in the destination */ 
00105         *px = *pIn++; 
00106  
00107         /* Update the pointer px to point to the next row of the transposed matrix */ 
00108         px += nRows; 
00109  
00110         /* Read and store the input element in the destination */ 
00111         *px = *pIn++; 
00112  
00113         /* Update the pointer px to point to the next row of the transposed matrix */ 
00114         px += nRows; 
00115  
00116         /* Read and store the input element in the destination */ 
00117         *px = *pIn++; 
00118  
00119         /* Update the pointer px to point to the next row of the transposed matrix */ 
00120         px += nRows; 
00121  
00122         /* Decrement the column loop counter */ 
00123         blkCnt--; 
00124       } 
00125  
00126       /* Perform matrix transpose for last 3 samples here. */ 
00127       blkCnt = nColumns % 0x4u; 
00128  
00129       while(blkCnt > 0u) 
00130       { 
00131         /* Read and store the input element in the destination */ 
00132         *px = *pIn++; 
00133  
00134         /* Update the pointer px to point to the next row of the transposed matrix */ 
00135         px += nRows; 
00136  
00137         /* Decrement the column loop counter */ 
00138         blkCnt--; 
00139       } 
00140  
00141       i++; 
00142  
00143       /* Decrement the row loop counter */ 
00144       row--; 
00145  
00146     } while(row > 0u);          /* row loop end  */ 
00147  
00148     /* Set status as ARM_MATH_SUCCESS */ 
00149     status = ARM_MATH_SUCCESS; 
00150   } 
00151  
00152   /* Return to application */ 
00153   return (status); 
00154 } 
00155  
00156 /**  
00157  * @} end of MatrixTrans group  
00158  */