1 /* ----------------------------------------------------------------------
2 * Copyright (C) 2010 ARM Limited. All rights reserved.
7 * Project: CMSIS DSP Library
8 * Title: arm_scale_q15.c
10 * Description: Multiplies a Q15 vector by a scalar.
12 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
14 * Version 1.0.10 2011/7/15
15 * Big Endian support added and Merged M0 and M3/M4 Source code.
17 * Version 1.0.3 2010/11/29
18 * Re-organized the CMSIS folders and updated documentation.
20 * Version 1.0.2 2010/11/11
21 * Documentation updated.
23 * Version 1.0.1 2010/10/05
24 * Production release and review comments incorporated.
26 * Version 1.0.0 2010/09/20
27 * Production release and review comments incorporated
29 * Version 0.0.7 2010/06/10
30 * Misra-C changes done
31 * -------------------------------------------------------------------- */
45 * @brief Multiplies a Q15 vector by a scalar.
46 * @param[in] *pSrc points to the input vector
47 * @param[in] scaleFract fractional portion of the scale value
48 * @param[in] shift number of bits to shift the result by
49 * @param[out] *pDst points to the output vector
50 * @param[in] blockSize number of samples in the vector
53 * <b>Scaling and Overflow Behavior:</b>
55 * The input data <code>*pSrc</code> and <code>scaleFract</code> are in 1.15 format.
56 * These are multiplied to yield a 2.30 intermediate result and this is shifted with saturation to 1.15 format.
67 int8_t kShift = 15 - shift; /* shift to apply after scaling */
68 uint32_t blkCnt; /* loop counter */
72 /* Run the below code for Cortex-M4 and Cortex-M3 */
74 q15_t in1, in2; /* Temporary variables */
78 blkCnt = blockSize >> 2u;
80 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
81 ** a second loop below computes the remaining 1 to 3 samples. */
84 /* Reading 2 inputs from memory */
88 /* Scale the inputs and then store the 2 results in the destination buffer
89 * in single cycle by packing the outputs */
90 #ifndef ARM_MATH_BIG_ENDIAN
93 __PKHBT(__SSAT((in1 * scaleFract) >> kShift, 16),
94 __SSAT((in2 * scaleFract) >> kShift, 16), 16);
99 __PKHBT(__SSAT((in2 * scaleFract) >> kShift, 16),
100 __SSAT((in1 * scaleFract) >> kShift, 16), 16);
102 #endif /* #ifndef ARM_MATH_BIG_ENDIAN */
107 #ifndef ARM_MATH_BIG_ENDIAN
110 __PKHBT(__SSAT((in1 * scaleFract) >> kShift, 16),
111 __SSAT((in2 * scaleFract) >> kShift, 16), 16);
116 __PKHBT(__SSAT((in2 * scaleFract) >> kShift, 16),
117 __SSAT((in1 * scaleFract) >> kShift, 16), 16);
119 #endif /* #ifndef ARM_MATH_BIG_ENDIAN */
121 /* Decrement the loop counter */
125 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
126 ** No loop unrolling is used. */
127 blkCnt = blockSize % 0x4u;
132 /* Scale the input and then store the result in the destination buffer. */
133 *pDst++ = (q15_t) (__SSAT(((*pSrc++) * scaleFract) >> kShift, 16));
135 /* Decrement the loop counter */
141 /* Run the below code for Cortex-M0 */
143 /* Initialize blkCnt with number of samples */
149 /* Scale the input and then store the result in the destination buffer. */
150 *pDst++ = (q15_t) (__SSAT(((q31_t) * pSrc++ * scaleFract) >> kShift, 16));
152 /* Decrement the loop counter */
156 #endif /* #ifndef ARM_MATH_CM0 */
161 * @} end of scale group