Welcome to mirror list, hosted at ThFree Co, Russian Federation.

github.com/Flipper-Zero/STM32CubeWB.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
path: root/Drivers/CMSIS/DSP/Source/ComplexMathFunctions/arm_cmplx_mult_cmplx_q31.c
diff options
context:
space:
mode:
Diffstat (limited to 'Drivers/CMSIS/DSP/Source/ComplexMathFunctions/arm_cmplx_mult_cmplx_q31.c')
-rw-r--r--Drivers/CMSIS/DSP/Source/ComplexMathFunctions/arm_cmplx_mult_cmplx_q31.c273
1 files changed, 48 insertions, 225 deletions
diff --git a/Drivers/CMSIS/DSP/Source/ComplexMathFunctions/arm_cmplx_mult_cmplx_q31.c b/Drivers/CMSIS/DSP/Source/ComplexMathFunctions/arm_cmplx_mult_cmplx_q31.c
index b01c4f675..f6d6dc6d5 100644
--- a/Drivers/CMSIS/DSP/Source/ComplexMathFunctions/arm_cmplx_mult_cmplx_q31.c
+++ b/Drivers/CMSIS/DSP/Source/ComplexMathFunctions/arm_cmplx_mult_cmplx_q31.c
@@ -3,13 +3,13 @@
* Title: arm_cmplx_mult_cmplx_q31.c
* Description: Q31 complex-by-complex multiplication
*
- * $Date: 27. January 2017
- * $Revision: V.1.5.1
+ * $Date: 18. March 2019
+ * $Revision: V1.6.0
*
* Target Processor: Cortex-M cores
* -------------------------------------------------------------------- */
/*
- * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ * Copyright (C) 2010-2019 ARM Limited or its affiliates. All rights reserved.
*
* SPDX-License-Identifier: Apache-2.0
*
@@ -29,286 +29,109 @@
#include "arm_math.h"
/**
- * @ingroup groupCmplxMath
+ @ingroup groupCmplxMath
*/
/**
- * @addtogroup CmplxByCmplxMult
- * @{
+ @addtogroup CmplxByCmplxMult
+ @{
*/
-
/**
- * @brief Q31 complex-by-complex multiplication
- * @param[in] *pSrcA points to the first input vector
- * @param[in] *pSrcB points to the second input vector
- * @param[out] *pDst points to the output vector
- * @param[in] numSamples number of complex samples in each vector
- * @return none.
- *
- * <b>Scaling and Overflow Behavior:</b>
- * \par
- * The function implements 1.31 by 1.31 multiplications and finally output is converted into 3.29 format.
- * Input down scaling is not required.
+ @brief Q31 complex-by-complex multiplication.
+ @param[in] pSrcA points to first input vector
+ @param[in] pSrcB points to second input vector
+ @param[out] pDst points to output vector
+ @param[in] numSamples number of samples in each vector
+ @return none
+
+ @par Scaling and Overflow Behavior
+ The function implements 1.31 by 1.31 multiplications and finally output is converted into 3.29 format.
+ Input down scaling is not required.
*/
void arm_cmplx_mult_cmplx_q31(
- q31_t * pSrcA,
- q31_t * pSrcB,
- q31_t * pDst,
- uint32_t numSamples)
+ const q31_t * pSrcA,
+ const q31_t * pSrcB,
+ q31_t * pDst,
+ uint32_t numSamples)
{
- q31_t a, b, c, d; /* Temporary variables to store real and imaginary values */
- uint32_t blkCnt; /* loop counters */
- q31_t mul1, mul2, mul3, mul4;
- q31_t out1, out2;
+ uint32_t blkCnt; /* Loop counter */
+ q31_t a, b, c, d; /* Temporary variables */
-#if defined (ARM_MATH_DSP)
+#if defined (ARM_MATH_LOOPUNROLL)
- /* Run the below code for Cortex-M4 and Cortex-M3 */
-
- /* loop Unrolling */
+ /* Loop unrolling: Compute 4 outputs at a time */
blkCnt = numSamples >> 2U;
- /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
- ** a second loop below computes the remaining 1 to 3 samples. */
while (blkCnt > 0U)
{
- /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
- /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
+ /* C[2 * i ] = A[2 * i] * B[2 * i ] - A[2 * i + 1] * B[2 * i + 1]. */
+ /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i ]. */
+
a = *pSrcA++;
b = *pSrcA++;
c = *pSrcB++;
d = *pSrcB++;
-
- mul1 = (q31_t) (((q63_t) a * c) >> 32);
- mul2 = (q31_t) (((q63_t) b * d) >> 32);
- mul3 = (q31_t) (((q63_t) a * d) >> 32);
- mul4 = (q31_t) (((q63_t) b * c) >> 32);
-
- mul1 = (mul1 >> 1);
- mul2 = (mul2 >> 1);
- mul3 = (mul3 >> 1);
- mul4 = (mul4 >> 1);
-
- out1 = mul1 - mul2;
- out2 = mul3 + mul4;
-
- /* store the real result in 3.29 format in the destination buffer. */
- *pDst++ = out1;
- /* store the imag result in 3.29 format in the destination buffer. */
- *pDst++ = out2;
+ /* store result in 3.29 format in destination buffer. */
+ *pDst++ = (q31_t) ( (((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33) );
+ *pDst++ = (q31_t) ( (((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33) );
a = *pSrcA++;
b = *pSrcA++;
c = *pSrcB++;
d = *pSrcB++;
-
- mul1 = (q31_t) (((q63_t) a * c) >> 32);
- mul2 = (q31_t) (((q63_t) b * d) >> 32);
- mul3 = (q31_t) (((q63_t) a * d) >> 32);
- mul4 = (q31_t) (((q63_t) b * c) >> 32);
-
- mul1 = (mul1 >> 1);
- mul2 = (mul2 >> 1);
- mul3 = (mul3 >> 1);
- mul4 = (mul4 >> 1);
-
- out1 = mul1 - mul2;
- out2 = mul3 + mul4;
-
- /* store the real result in 3.29 format in the destination buffer. */
- *pDst++ = out1;
- /* store the imag result in 3.29 format in the destination buffer. */
- *pDst++ = out2;
+ *pDst++ = (q31_t) ( (((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33) );
+ *pDst++ = (q31_t) ( (((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33) );
a = *pSrcA++;
b = *pSrcA++;
c = *pSrcB++;
d = *pSrcB++;
-
- mul1 = (q31_t) (((q63_t) a * c) >> 32);
- mul2 = (q31_t) (((q63_t) b * d) >> 32);
- mul3 = (q31_t) (((q63_t) a * d) >> 32);
- mul4 = (q31_t) (((q63_t) b * c) >> 32);
-
- mul1 = (mul1 >> 1);
- mul2 = (mul2 >> 1);
- mul3 = (mul3 >> 1);
- mul4 = (mul4 >> 1);
-
- out1 = mul1 - mul2;
- out2 = mul3 + mul4;
-
- /* store the real result in 3.29 format in the destination buffer. */
- *pDst++ = out1;
- /* store the imag result in 3.29 format in the destination buffer. */
- *pDst++ = out2;
+ *pDst++ = (q31_t) ( (((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33) );
+ *pDst++ = (q31_t) ( (((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33) );
a = *pSrcA++;
b = *pSrcA++;
c = *pSrcB++;
d = *pSrcB++;
+ *pDst++ = (q31_t) ( (((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33) );
+ *pDst++ = (q31_t) ( (((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33) );
- mul1 = (q31_t) (((q63_t) a * c) >> 32);
- mul2 = (q31_t) (((q63_t) b * d) >> 32);
- mul3 = (q31_t) (((q63_t) a * d) >> 32);
- mul4 = (q31_t) (((q63_t) b * c) >> 32);
-
- mul1 = (mul1 >> 1);
- mul2 = (mul2 >> 1);
- mul3 = (mul3 >> 1);
- mul4 = (mul4 >> 1);
-
- out1 = mul1 - mul2;
- out2 = mul3 + mul4;
-
- /* store the real result in 3.29 format in the destination buffer. */
- *pDst++ = out1;
- /* store the imag result in 3.29 format in the destination buffer. */
- *pDst++ = out2;
-
- /* Decrement the blockSize loop counter */
+ /* Decrement loop counter */
blkCnt--;
}
- /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
- ** No loop unrolling is used. */
+ /* Loop unrolling: Compute remaining outputs */
blkCnt = numSamples % 0x4U;
- while (blkCnt > 0U)
- {
- /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
- /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
- a = *pSrcA++;
- b = *pSrcA++;
- c = *pSrcB++;
- d = *pSrcB++;
-
- mul1 = (q31_t) (((q63_t) a * c) >> 32);
- mul2 = (q31_t) (((q63_t) b * d) >> 32);
- mul3 = (q31_t) (((q63_t) a * d) >> 32);
- mul4 = (q31_t) (((q63_t) b * c) >> 32);
-
- mul1 = (mul1 >> 1);
- mul2 = (mul2 >> 1);
- mul3 = (mul3 >> 1);
- mul4 = (mul4 >> 1);
-
- out1 = mul1 - mul2;
- out2 = mul3 + mul4;
-
- /* store the real result in 3.29 format in the destination buffer. */
- *pDst++ = out1;
- /* store the imag result in 3.29 format in the destination buffer. */
- *pDst++ = out2;
-
- /* Decrement the blockSize loop counter */
- blkCnt--;
- }
-
#else
- /* Run the below code for Cortex-M0 */
+ /* Initialize blkCnt with number of samples */
+ blkCnt = numSamples;
- /* loop Unrolling */
- blkCnt = numSamples >> 1U;
+#endif /* #if defined (ARM_MATH_LOOPUNROLL) */
- /* First part of the processing with loop unrolling. Compute 2 outputs at a time.
- ** a second loop below computes the remaining 1 sample. */
while (blkCnt > 0U)
{
- /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
- /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
- a = *pSrcA++;
- b = *pSrcA++;
- c = *pSrcB++;
- d = *pSrcB++;
-
- mul1 = (q31_t) (((q63_t) a * c) >> 32);
- mul2 = (q31_t) (((q63_t) b * d) >> 32);
- mul3 = (q31_t) (((q63_t) a * d) >> 32);
- mul4 = (q31_t) (((q63_t) b * c) >> 32);
-
- mul1 = (mul1 >> 1);
- mul2 = (mul2 >> 1);
- mul3 = (mul3 >> 1);
- mul4 = (mul4 >> 1);
-
- out1 = mul1 - mul2;
- out2 = mul3 + mul4;
-
- /* store the real result in 3.29 format in the destination buffer. */
- *pDst++ = out1;
- /* store the imag result in 3.29 format in the destination buffer. */
- *pDst++ = out2;
-
- a = *pSrcA++;
- b = *pSrcA++;
- c = *pSrcB++;
- d = *pSrcB++;
-
- mul1 = (q31_t) (((q63_t) a * c) >> 32);
- mul2 = (q31_t) (((q63_t) b * d) >> 32);
- mul3 = (q31_t) (((q63_t) a * d) >> 32);
- mul4 = (q31_t) (((q63_t) b * c) >> 32);
-
- mul1 = (mul1 >> 1);
- mul2 = (mul2 >> 1);
- mul3 = (mul3 >> 1);
- mul4 = (mul4 >> 1);
-
- out1 = mul1 - mul2;
- out2 = mul3 + mul4;
-
- /* store the real result in 3.29 format in the destination buffer. */
- *pDst++ = out1;
- /* store the imag result in 3.29 format in the destination buffer. */
- *pDst++ = out2;
-
- /* Decrement the blockSize loop counter */
- blkCnt--;
- }
+ /* C[2 * i ] = A[2 * i] * B[2 * i ] - A[2 * i + 1] * B[2 * i + 1]. */
+ /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i ]. */
- /* If the blockSize is not a multiple of 2, compute any remaining output samples here.
- ** No loop unrolling is used. */
- blkCnt = numSamples % 0x2U;
-
- while (blkCnt > 0U)
- {
- /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1]. */
- /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i]. */
a = *pSrcA++;
b = *pSrcA++;
c = *pSrcB++;
d = *pSrcB++;
- mul1 = (q31_t) (((q63_t) a * c) >> 32);
- mul2 = (q31_t) (((q63_t) b * d) >> 32);
- mul3 = (q31_t) (((q63_t) a * d) >> 32);
- mul4 = (q31_t) (((q63_t) b * c) >> 32);
-
- mul1 = (mul1 >> 1);
- mul2 = (mul2 >> 1);
- mul3 = (mul3 >> 1);
- mul4 = (mul4 >> 1);
+ /* store result in 3.29 format in destination buffer. */
+ *pDst++ = (q31_t) ( (((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33) );
+ *pDst++ = (q31_t) ( (((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33) );
- out1 = mul1 - mul2;
- out2 = mul3 + mul4;
-
- /* store the real result in 3.29 format in the destination buffer. */
- *pDst++ = out1;
- /* store the imag result in 3.29 format in the destination buffer. */
- *pDst++ = out2;
-
- /* Decrement the blockSize loop counter */
+ /* Decrement loop counter */
blkCnt--;
}
-#endif /* #if defined (ARM_MATH_DSP) */
-
}
/**
- * @} end of CmplxByCmplxMult group
+ @} end of CmplxByCmplxMult group
*/
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-04 01:07:02 +0300