diff options
author | Jack Humbert <jack.humb@gmail.com> | 2016-06-18 14:30:24 -0400 |
---|---|---|
committer | GitHub <noreply@github.com> | 2016-06-18 14:30:24 -0400 |
commit | db32864ce7029d758f57729cc2f75e051a28d0a2 (patch) | |
tree | f3ac60d9c826a9ad5ef5bc4d199efaddae156ba6 /tmk_core/tool/mbed/mbed-sdk/libraries/dsp/cmsis_dsp/MatrixFunctions/arm_mat_mult_q15.c | |
parent | 1923cffd41d9d71cd9f434092654dba05513137b (diff) | |
download | firmware-db32864ce7029d758f57729cc2f75e051a28d0a2.tar.gz firmware-db32864ce7029d758f57729cc2f75e051a28d0a2.tar.bz2 firmware-db32864ce7029d758f57729cc2f75e051a28d0a2.zip |
Cleans up quantum/keymap situation, removes extra lufa folders (#416)
* sorts out keycodes
* move midi around
* remove mbed
* replaces keymap with qmk/keymap_common
* fixes keymap.h
* keymap, config, quantum rearrange
* removes unneeded lufa stuff
Diffstat (limited to 'tmk_core/tool/mbed/mbed-sdk/libraries/dsp/cmsis_dsp/MatrixFunctions/arm_mat_mult_q15.c')
-rw-r--r-- | tmk_core/tool/mbed/mbed-sdk/libraries/dsp/cmsis_dsp/MatrixFunctions/arm_mat_mult_q15.c | 469 |
1 files changed, 0 insertions, 469 deletions
diff --git a/tmk_core/tool/mbed/mbed-sdk/libraries/dsp/cmsis_dsp/MatrixFunctions/arm_mat_mult_q15.c b/tmk_core/tool/mbed/mbed-sdk/libraries/dsp/cmsis_dsp/MatrixFunctions/arm_mat_mult_q15.c deleted file mode 100644 index 1e112ab77..000000000 --- a/tmk_core/tool/mbed/mbed-sdk/libraries/dsp/cmsis_dsp/MatrixFunctions/arm_mat_mult_q15.c +++ /dev/null @@ -1,469 +0,0 @@ -/* ---------------------------------------------------------------------- -* Copyright (C) 2010-2013 ARM Limited. All rights reserved. -* -* $Date: 17. January 2013 -* $Revision: V1.4.1 -* -* Project: CMSIS DSP Library -* Title: arm_mat_mult_q15.c -* -* Description: Q15 matrix multiplication. -* -* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0 -* -* Redistribution and use in source and binary forms, with or without -* modification, are permitted provided that the following conditions -* are met: -* - Redistributions of source code must retain the above copyright -* notice, this list of conditions and the following disclaimer. -* - Redistributions in binary form must reproduce the above copyright -* notice, this list of conditions and the following disclaimer in -* the documentation and/or other materials provided with the -* distribution. -* - Neither the name of ARM LIMITED nor the names of its contributors -* may be used to endorse or promote products derived from this -* software without specific prior written permission. -* -* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS -* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT -* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS -* FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE -* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, -* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, -* BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; -* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT -* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN -* ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -* POSSIBILITY OF SUCH DAMAGE. -* -------------------------------------------------------------------- */ - -#include "arm_math.h" - -/** - * @ingroup groupMatrix - */ - -/** - * @addtogroup MatrixMult - * @{ - */ - - -/** - * @brief Q15 matrix multiplication - * @param[in] *pSrcA points to the first input matrix structure - * @param[in] *pSrcB points to the second input matrix structure - * @param[out] *pDst points to output matrix structure - * @param[in] *pState points to the array for storing intermediate results - * @return The function returns either - * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking. - * - * @details - * <b>Scaling and Overflow Behavior:</b> - * - * \par - * The function is implemented using a 64-bit internal accumulator. The inputs to the - * multiplications are in 1.15 format and multiplications yield a 2.30 result. - * The 2.30 intermediate - * results are accumulated in a 64-bit accumulator in 34.30 format. This approach - * provides 33 guard bits and there is no risk of overflow. The 34.30 result is then - * truncated to 34.15 format by discarding the low 15 bits and then saturated to - * 1.15 format. - * - * \par - * Refer to <code>arm_mat_mult_fast_q15()</code> for a faster but less precise version of this function for Cortex-M3 and Cortex-M4. - * - */ - -arm_status arm_mat_mult_q15( - const arm_matrix_instance_q15 * pSrcA, - const arm_matrix_instance_q15 * pSrcB, - arm_matrix_instance_q15 * pDst, - q15_t * pState CMSIS_UNUSED) -{ - q63_t sum; /* accumulator */ - -#ifndef ARM_MATH_CM0_FAMILY - - /* Run the below code for Cortex-M4 and Cortex-M3 */ - - q15_t *pSrcBT = pState; /* input data matrix pointer for transpose */ - q15_t *pInA = pSrcA->pData; /* input data matrix pointer A of Q15 type */ - q15_t *pInB = pSrcB->pData; /* input data matrix pointer B of Q15 type */ - q15_t *px; /* Temporary output data matrix pointer */ - uint16_t numRowsA = pSrcA->numRows; /* number of rows of input matrix A */ - uint16_t numColsB = pSrcB->numCols; /* number of columns of input matrix B */ - uint16_t numColsA = pSrcA->numCols; /* number of columns of input matrix A */ - uint16_t numRowsB = pSrcB->numRows; /* number of rows of input matrix A */ - uint16_t col, i = 0u, row = numRowsB, colCnt; /* loop counters */ - arm_status status; /* status of matrix multiplication */ - -#ifndef UNALIGNED_SUPPORT_DISABLE - - q31_t in; /* Temporary variable to hold the input value */ - q31_t pSourceA1, pSourceB1, pSourceA2, pSourceB2; - -#else - - q15_t in; /* Temporary variable to hold the input value */ - q15_t inA1, inB1, inA2, inB2; - -#endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */ - -#ifdef ARM_MATH_MATRIX_CHECK - /* Check for matrix mismatch condition */ - if((pSrcA->numCols != pSrcB->numRows) || - (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols)) - { - /* Set status as ARM_MATH_SIZE_MISMATCH */ - status = ARM_MATH_SIZE_MISMATCH; - } - else -#endif /* #ifdef ARM_MATH_MATRIX_CHECK */ - { - /* Matrix transpose */ - do - { - /* Apply loop unrolling and exchange the columns with row elements */ - col = numColsB >> 2; - - /* The pointer px is set to starting address of the column being processed */ - px = pSrcBT + i; - - /* First part of the processing with loop unrolling. Compute 4 outputs at a time. - ** a second loop below computes the remaining 1 to 3 samples. */ - while(col > 0u) - { -#ifndef UNALIGNED_SUPPORT_DISABLE - - /* Read two elements from the row */ - in = *__SIMD32(pInB)++; - - /* Unpack and store one element in the destination */ -#ifndef ARM_MATH_BIG_ENDIAN - - *px = (q15_t) in; - -#else - - *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16); - -#endif /* #ifndef ARM_MATH_BIG_ENDIAN */ - - /* Update the pointer px to point to the next row of the transposed matrix */ - px += numRowsB; - - /* Unpack and store the second element in the destination */ -#ifndef ARM_MATH_BIG_ENDIAN - - *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16); - -#else - - *px = (q15_t) in; - -#endif /* #ifndef ARM_MATH_BIG_ENDIAN */ - - /* Update the pointer px to point to the next row of the transposed matrix */ - px += numRowsB; - - /* Read two elements from the row */ - in = *__SIMD32(pInB)++; - - /* Unpack and store one element in the destination */ -#ifndef ARM_MATH_BIG_ENDIAN - - *px = (q15_t) in; - -#else - - *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16); - -#endif /* #ifndef ARM_MATH_BIG_ENDIAN */ - - /* Update the pointer px to point to the next row of the transposed matrix */ - px += numRowsB; - - /* Unpack and store the second element in the destination */ - -#ifndef ARM_MATH_BIG_ENDIAN - - *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16); - -#else - - *px = (q15_t) in; - -#endif /* #ifndef ARM_MATH_BIG_ENDIAN */ - - /* Update the pointer px to point to the next row of the transposed matrix */ - px += numRowsB; - -#else - - /* Read one element from the row */ - in = *pInB++; - - /* Store one element in the destination */ - *px = in; - - /* Update the pointer px to point to the next row of the transposed matrix */ - px += numRowsB; - - /* Read one element from the row */ - in = *pInB++; - - /* Store one element in the destination */ - *px = in; - - /* Update the pointer px to point to the next row of the transposed matrix */ - px += numRowsB; - - /* Read one element from the row */ - in = *pInB++; - - /* Store one element in the destination */ - *px = in; - - /* Update the pointer px to point to the next row of the transposed matrix */ - px += numRowsB; - - /* Read one element from the row */ - in = *pInB++; - - /* Store one element in the destination */ - *px = in; - - /* Update the pointer px to point to the next row of the transposed matrix */ - px += numRowsB; - -#endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */ - - /* Decrement the column loop counter */ - col--; - } - - /* If the columns of pSrcB is not a multiple of 4, compute any remaining output samples here. - ** No loop unrolling is used. */ - col = numColsB % 0x4u; - - while(col > 0u) - { - /* Read and store the input element in the destination */ - *px = *pInB++; - - /* Update the pointer px to point to the next row of the transposed matrix */ - px += numRowsB; - - /* Decrement the column loop counter */ - col--; - } - - i++; - - /* Decrement the row loop counter */ - row--; - - } while(row > 0u); - - /* Reset the variables for the usage in the following multiplication process */ - row = numRowsA; - i = 0u; - px = pDst->pData; - - /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */ - /* row loop */ - do - { - /* For every row wise process, the column loop counter is to be initiated */ - col = numColsB; - - /* For every row wise process, the pIn2 pointer is set - ** to the starting address of the transposed pSrcB data */ - pInB = pSrcBT; - - /* column loop */ - do - { - /* Set the variable sum, that acts as accumulator, to zero */ - sum = 0; - - /* Apply loop unrolling and compute 2 MACs simultaneously. */ - colCnt = numColsA >> 2; - - /* Initiate the pointer pIn1 to point to the starting address of the column being processed */ - pInA = pSrcA->pData + i; - - - /* matrix multiplication */ - while(colCnt > 0u) - { - /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */ -#ifndef UNALIGNED_SUPPORT_DISABLE - - /* read real and imag values from pSrcA and pSrcB buffer */ - pSourceA1 = *__SIMD32(pInA)++; - pSourceB1 = *__SIMD32(pInB)++; - - pSourceA2 = *__SIMD32(pInA)++; - pSourceB2 = *__SIMD32(pInB)++; - - /* Multiply and Accumlates */ - sum = __SMLALD(pSourceA1, pSourceB1, sum); - sum = __SMLALD(pSourceA2, pSourceB2, sum); - -#else - /* read real and imag values from pSrcA and pSrcB buffer */ - inA1 = *pInA++; - inB1 = *pInB++; - inA2 = *pInA++; - /* Multiply and Accumlates */ - sum += inA1 * inB1; - inB2 = *pInB++; - - inA1 = *pInA++; - inB1 = *pInB++; - /* Multiply and Accumlates */ - sum += inA2 * inB2; - inA2 = *pInA++; - inB2 = *pInB++; - - /* Multiply and Accumlates */ - sum += inA1 * inB1; - sum += inA2 * inB2; - -#endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */ - - /* Decrement the loop counter */ - colCnt--; - } - - /* process remaining column samples */ - colCnt = numColsA & 3u; - - while(colCnt > 0u) - { - /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */ - sum += *pInA++ * *pInB++; - - /* Decrement the loop counter */ - colCnt--; - } - - /* Saturate and store the result in the destination buffer */ - *px = (q15_t) (__SSAT((sum >> 15), 16)); - px++; - - /* Decrement the column loop counter */ - col--; - - } while(col > 0u); - - i = i + numColsA; - - /* Decrement the row loop counter */ - row--; - - } while(row > 0u); - -#else - - /* Run the below code for Cortex-M0 */ - - q15_t *pIn1 = pSrcA->pData; /* input data matrix pointer A */ - q15_t *pIn2 = pSrcB->pData; /* input data matrix pointer B */ - q15_t *pInA = pSrcA->pData; /* input data matrix pointer A of Q15 type */ - q15_t *pInB = pSrcB->pData; /* input data matrix pointer B of Q15 type */ - q15_t *pOut = pDst->pData; /* output data matrix pointer */ - q15_t *px; /* Temporary output data matrix pointer */ - uint16_t numColsB = pSrcB->numCols; /* number of columns of input matrix B */ - uint16_t numColsA = pSrcA->numCols; /* number of columns of input matrix A */ - uint16_t numRowsA = pSrcA->numRows; /* number of rows of input matrix A */ - uint16_t col, i = 0u, row = numRowsA, colCnt; /* loop counters */ - arm_status status; /* status of matrix multiplication */ - -#ifdef ARM_MATH_MATRIX_CHECK - - /* Check for matrix mismatch condition */ - if((pSrcA->numCols != pSrcB->numRows) || - (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols)) - { - /* Set status as ARM_MATH_SIZE_MISMATCH */ - status = ARM_MATH_SIZE_MISMATCH; - } - else -#endif /* #ifdef ARM_MATH_MATRIX_CHECK */ - - { - /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */ - /* row loop */ - do - { - /* Output pointer is set to starting address of the row being processed */ - px = pOut + i; - - /* For every row wise process, the column loop counter is to be initiated */ - col = numColsB; - - /* For every row wise process, the pIn2 pointer is set - ** to the starting address of the pSrcB data */ - pIn2 = pSrcB->pData; - - /* column loop */ - do - { - /* Set the variable sum, that acts as accumulator, to zero */ - sum = 0; - - /* Initiate the pointer pIn1 to point to the starting address of pSrcA */ - pIn1 = pInA; - - /* Matrix A columns number of MAC operations are to be performed */ - colCnt = numColsA; - - /* matrix multiplication */ - while(colCnt > 0u) - { - /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */ - /* Perform the multiply-accumulates */ - sum += (q31_t) * pIn1++ * *pIn2; - pIn2 += numColsB; - - /* Decrement the loop counter */ - colCnt--; - } - - /* Convert the result from 34.30 to 1.15 format and store the saturated value in destination buffer */ - /* Saturate and store the result in the destination buffer */ - *px++ = (q15_t) __SSAT((sum >> 15), 16); - - /* Decrement the column loop counter */ - col--; - - /* Update the pointer pIn2 to point to the starting address of the next column */ - pIn2 = pInB + (numColsB - col); - - } while(col > 0u); - - /* Update the pointer pSrcA to point to the starting address of the next row */ - i = i + numColsB; - pInA = pInA + numColsA; - - /* Decrement the row loop counter */ - row--; - - } while(row > 0u); - -#endif /* #ifndef ARM_MATH_CM0_FAMILY */ - /* set status as ARM_MATH_SUCCESS */ - status = ARM_MATH_SUCCESS; - } - - /* Return to application */ - return (status); -} - -/** - * @} end of MatrixMult group - */ |