]>
git.gir.st - tmk_keyboard.git/blob - tmk_core/tool/mbed/mbed-sdk/libraries/dsp/cmsis_dsp/BasicMathFunctions/arm_dot_prod_q7.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_q7.c
10 * Description: Q7 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 Q7 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 1.7 x 1.7 = 2.14 format and these
63 * results are added to an accumulator in 18.14 format.
64 * Nonsaturating additions are used and there is no danger of wrap around as long as
65 * the vectors are less than 2^18 elements long.
66 * The return result is in 18.14 format.
75 uint32_t blkCnt
; /* loop counter */
77 q31_t sum
= 0; /* Temporary variables to store output */
79 #ifndef ARM_MATH_CM0_FAMILY
81 /* Run the below code for Cortex-M4 and Cortex-M3 */
83 q31_t input1
, input2
; /* Temporary variables to store input */
84 q31_t inA1
, inA2
, inB1
, inB2
; /* Temporary variables to store input */
89 blkCnt
= blockSize
>> 2u;
91 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
92 ** a second loop below computes the remaining 1 to 3 samples. */
95 /* read 4 samples at a time from sourceA */
96 input1
= *__SIMD32(pSrcA
)++;
97 /* read 4 samples at a time from sourceB */
98 input2
= *__SIMD32(pSrcB
)++;
100 /* extract two q7_t samples to q15_t samples */
101 inA1
= __SXTB16(__ROR(input1
, 8));
102 /* extract reminaing two samples */
103 inA2
= __SXTB16(input1
);
104 /* extract two q7_t samples to q15_t samples */
105 inB1
= __SXTB16(__ROR(input2
, 8));
106 /* extract reminaing two samples */
107 inB2
= __SXTB16(input2
);
109 /* multiply and accumulate two samples at a time */
110 sum
= __SMLAD(inA1
, inB1
, sum
);
111 sum
= __SMLAD(inA2
, inB2
, sum
);
113 /* Decrement the loop counter */
117 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
118 ** No loop unrolling is used. */
119 blkCnt
= blockSize
% 0x4u
;
123 /* C = A[0]* B[0] + A[1]* B[1] + A[2]* B[2] + .....+ A[blockSize-1]* B[blockSize-1] */
124 /* Dot product and then store the results in a temporary buffer. */
125 sum
= __SMLAD(*pSrcA
++, *pSrcB
++, sum
);
127 /* Decrement the loop counter */
133 /* Run the below code for Cortex-M0 */
137 /* Initialize blkCnt with number of samples */
142 /* C = A[0]* B[0] + A[1]* B[1] + A[2]* B[2] + .....+ A[blockSize-1]* B[blockSize-1] */
143 /* Dot product and then store the results in a temporary buffer. */
144 sum
+= (q31_t
) ((q15_t
) * pSrcA
++ * *pSrcB
++);
146 /* Decrement the loop counter */
150 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
153 /* Store the result in the destination buffer in 18.14 format */
158 * @} end of dot_prod group