1 /* ----------------------------------------------------------------------
2 * Copyright (C) 2010-2013 ARM Limited. All rights reserved.
4 * $Date: 17. January 2013
7 * Project: CMSIS DSP Library
8 * Title: arm_shift_q31.c
10 * Description: Shifts the elements of a Q31 vector by a specified number of bits.
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 * -------------------------------------------------------------------- */
47 * @defgroup shift Vector Shift
49 * Shifts the elements of a fixed-point vector by a specified number of bits.
50 * There are separate functions for Q7, Q15, and Q31 data types.
51 * The underlying algorithm used is:
54 * pDst[n] = pSrc[n] << shift, 0 <= n < blockSize.
57 * If <code>shift</code> is positive then the elements of the vector are shifted to the left.
58 * If <code>shift</code> is negative then the elements of the vector are shifted to the right.
60 * The functions support in-place computation allowing the source and destination
61 * pointers to reference the same memory buffer.
70 * @brief Shifts the elements of a Q31 vector a specified number of bits.
71 * @param[in] *pSrc points to the input vector
72 * @param[in] shiftBits number of bits to shift. A positive value shifts left; a negative value shifts right.
73 * @param[out] *pDst points to the output vector
74 * @param[in] blockSize number of samples in the vector
78 * <b>Scaling and Overflow Behavior:</b>
80 * The function uses saturating arithmetic.
81 * Results outside of the allowable Q31 range [0x80000000 0x7FFFFFFF] will be saturated.
90 uint32_t blkCnt; /* loop counter */
91 uint8_t sign = (shiftBits & 0x80); /* Sign of shiftBits */
93 #ifndef ARM_MATH_CM0_FAMILY
95 q31_t in1, in2, in3, in4; /* Temporary input variables */
96 q31_t out1, out2, out3, out4; /* Temporary output variables */
99 blkCnt = blockSize >> 2u;
104 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
105 ** a second loop below computes the remaining 1 to 3 samples. */
108 /* C = A << shiftBits */
109 /* Shift the input and then store the results in the destination buffer. */
112 out1 = in1 << shiftBits;
114 out2 = in2 << shiftBits;
116 if(in1 != (out1 >> shiftBits))
117 out1 = 0x7FFFFFFF ^ (in1 >> 31);
119 if(in2 != (out2 >> shiftBits))
120 out2 = 0x7FFFFFFF ^ (in2 >> 31);
123 out3 = in3 << shiftBits;
125 out4 = in4 << shiftBits;
127 if(in3 != (out3 >> shiftBits))
128 out3 = 0x7FFFFFFF ^ (in3 >> 31);
130 if(in4 != (out4 >> shiftBits))
131 out4 = 0x7FFFFFFF ^ (in4 >> 31);
136 /* Update destination pointer to process next sampels */
140 /* Decrement the loop counter */
147 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
148 ** a second loop below computes the remaining 1 to 3 samples. */
151 /* C = A >> shiftBits */
152 /* Shift the input and then store the results in the destination buffer. */
158 *pDst = (in1 >> -shiftBits);
159 *(pDst + 1) = (in2 >> -shiftBits);
160 *(pDst + 2) = (in3 >> -shiftBits);
161 *(pDst + 3) = (in4 >> -shiftBits);
172 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
173 ** No loop unrolling is used. */
174 blkCnt = blockSize % 0x4u;
178 /* Run the below code for Cortex-M0 */
181 /* Initialize blkCnt with number of samples */
184 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
189 /* C = A (>> or <<) shiftBits */
190 /* Shift the input and then store the result in the destination buffer. */
191 *pDst++ = (sign == 0u) ? clip_q63_to_q31((q63_t) * pSrc++ << shiftBits) :
192 (*pSrc++ >> -shiftBits);
194 /* Decrement the loop counter */
202 * @} end of shift group