]>
git.gir.st - tmk_keyboard.git/blob - tmk_core/tool/mbed/mbed-sdk/libraries/dsp/cmsis_dsp/BasicMathFunctions/arm_dot_prod_q31.c
1 /* ----------------------------------------------------------------------
2 * Copyright (C) 2010-2013 ARM Limited. All rights reserved.
4 * $Date: 17. January 2013
7 * Project: CMSIS DSP Library
8 * Title: arm_dot_prod_q31.c
10 * Description: Q31 dot product.
12 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
14 * Redistribution and use in source and binary forms, with or without
15 * modification, are permitted provided that the following conditions
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
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.
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 * -------------------------------------------------------------------- */
48 * @addtogroup dot_prod
53 * @brief Dot product of Q31 vectors.
54 * @param[in] *pSrcA points to the first input vector
55 * @param[in] *pSrcB points to the second input vector
56 * @param[in] blockSize number of samples in each vector
57 * @param[out] *result output result returned here
60 * <b>Scaling and Overflow Behavior:</b>
62 * The intermediate multiplications are in 1.31 x 1.31 = 2.62 format and these
63 * are truncated to 2.48 format by discarding the lower 14 bits.
64 * The 2.48 result is then added without saturation to a 64-bit accumulator in 16.48 format.
65 * There are 15 guard bits in the accumulator and there is no risk of overflow as long as
66 * the length of the vectors is less than 2^16 elements.
67 * The return result is in 16.48 format.
70 void arm_dot_prod_q31(
76 q63_t sum
= 0; /* Temporary result storage */
77 uint32_t blkCnt
; /* loop counter */
80 #ifndef ARM_MATH_CM0_FAMILY
82 /* Run the below code for Cortex-M4 and Cortex-M3 */
83 q31_t inA1
, inA2
, inA3
, inA4
;
84 q31_t inB1
, inB2
, inB3
, inB4
;
87 blkCnt
= blockSize
>> 2u;
89 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
90 ** a second loop below computes the remaining 1 to 3 samples. */
93 /* C = A[0]* B[0] + A[1]* B[1] + A[2]* B[2] + .....+ A[blockSize-1]* B[blockSize-1] */
94 /* Calculate dot product and then store the result in a temporary buffer. */
104 sum
+= ((q63_t
) inA1
* inB1
) >> 14u;
105 sum
+= ((q63_t
) inA2
* inB2
) >> 14u;
106 sum
+= ((q63_t
) inA3
* inB3
) >> 14u;
107 sum
+= ((q63_t
) inA4
* inB4
) >> 14u;
109 /* Decrement the loop counter */
113 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
114 ** No loop unrolling is used. */
115 blkCnt
= blockSize
% 0x4u
;
119 /* Run the below code for Cortex-M0 */
121 /* Initialize blkCnt with number of samples */
124 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
129 /* C = A[0]* B[0] + A[1]* B[1] + A[2]* B[2] + .....+ A[blockSize-1]* B[blockSize-1] */
130 /* Calculate dot product and then store the result in a temporary buffer. */
131 sum
+= ((q63_t
) * pSrcA
++ * *pSrcB
++) >> 14u;
133 /* Decrement the loop counter */
137 /* Store the result in the destination buffer in 16.48 format */
142 * @} end of dot_prod group