1 /* ----------------------------------------------------------------------
2 * Copyright (C) 2010-2013 ARM Limited. All rights reserved.
4 * $Date: 17. January 2013
7 * Project: CMSIS DSP Library
8 * Title: arm_scale_q31.c
10 * Description: Multiplies a Q31 vector by a scalar.
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 * -------------------------------------------------------------------- */
53 * @brief Multiplies a Q31 vector by a scalar.
54 * @param[in] *pSrc points to the input vector
55 * @param[in] scaleFract fractional portion of the scale value
56 * @param[in] shift number of bits to shift the result by
57 * @param[out] *pDst points to the output vector
58 * @param[in] blockSize number of samples in the vector
61 * <b>Scaling and Overflow Behavior:</b>
63 * The input data <code>*pSrc</code> and <code>scaleFract</code> are in 1.31 format.
64 * These are multiplied to yield a 2.62 intermediate result and this is shifted with saturation to 1.31 format.
74 int8_t kShift = shift + 1; /* Shift to apply after scaling */
75 int8_t sign = (kShift & 0x80);
76 uint32_t blkCnt; /* loop counter */
79 #ifndef ARM_MATH_CM0_FAMILY
81 /* Run the below code for Cortex-M4 and Cortex-M3 */
83 q31_t in1, in2, in3, in4; /* temporary input variables */
84 q31_t out1, out2, out3, out4; /* temporary output variabels */
88 blkCnt = blockSize >> 2u;
92 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
93 ** a second loop below computes the remaining 1 to 3 samples. */
96 /* read four inputs from source */
102 /* multiply input with scaler value */
103 in1 = ((q63_t) in1 * scaleFract) >> 32;
104 in2 = ((q63_t) in2 * scaleFract) >> 32;
105 in3 = ((q63_t) in3 * scaleFract) >> 32;
106 in4 = ((q63_t) in4 * scaleFract) >> 32;
109 out1 = in1 << kShift;
110 out2 = in2 << kShift;
112 /* saturate the results. */
113 if(in1 != (out1 >> kShift))
114 out1 = 0x7FFFFFFF ^ (in1 >> 31);
116 if(in2 != (out2 >> kShift))
117 out2 = 0x7FFFFFFF ^ (in2 >> 31);
119 out3 = in3 << kShift;
120 out4 = in4 << kShift;
125 if(in3 != (out3 >> kShift))
126 out3 = 0x7FFFFFFF ^ (in3 >> 31);
128 if(in4 != (out4 >> kShift))
129 out4 = 0x7FFFFFFF ^ (in4 >> 31);
131 /* Store result destination */
135 /* Update pointers to process next sampels */
139 /* Decrement the loop counter */
146 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
147 ** a second loop below computes the remaining 1 to 3 samples. */
150 /* read four inputs from source */
156 /* multiply input with scaler value */
157 in1 = ((q63_t) in1 * scaleFract) >> 32;
158 in2 = ((q63_t) in2 * scaleFract) >> 32;
159 in3 = ((q63_t) in3 * scaleFract) >> 32;
160 in4 = ((q63_t) in4 * scaleFract) >> 32;
163 out1 = in1 >> -kShift;
164 out2 = in2 >> -kShift;
166 out3 = in3 >> -kShift;
167 out4 = in4 >> -kShift;
169 /* Store result destination */
176 /* Update pointers to process next sampels */
180 /* Decrement the loop counter */
184 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
185 ** No loop unrolling is used. */
186 blkCnt = blockSize % 0x4u;
190 /* Run the below code for Cortex-M0 */
192 /* Initialize blkCnt with number of samples */
195 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
202 /* Scale the input and then store the result in the destination buffer. */
204 in = ((q63_t) in * scaleFract) >> 32;
208 if(in != (out >> kShift))
209 out = 0x7FFFFFFF ^ (in >> 31);
213 /* Decrement the loop counter */
222 /* Scale the input and then store the result in the destination buffer. */
224 in = ((q63_t) in * scaleFract) >> 32;
230 /* Decrement the loop counter */
238 * @} end of scale group