Supported Simulink Blocks with CMSIS Library for ARM Cortex-A Processors

Embedded Coder^® Support Package for ARM^® Cortex^®-A Processors provides a code replacement library (CRL), ARM Cortex-A CMSIS, to generate calls to the CMSIS-DSP library optimized for ARM Cortex-A processors.

For more information on the CMSIS-DSP library, see https://github.com/ARM-software/CMSIS-DSP.

To improve the speed of your models and the performance of the generated code, you can set the model configuration parameters under Code Generation > Optimization. For more information on how to set the model parameters, see Model Configuration Parameters: Code Generation Optimization.

Note

ARM Cortex-A CMSIS CRL supports both 32-bit and 64-bit ARM Cortex-A architectures.
Only 64-bit ARM Cortex-A architectures support the half data type.
When the ARM Cortex-A CMSIS CRL does not provide code replacements because the input vector length being below the threshold, you can enable CMSIS function generation for all input lengths. To enable this option, set the environment variable DISABLE_ARMCORTEXA_CMSIS_CRL_THRESHOLDS to true using the following command:
```
setenv('DISABLE_ARMCORTEXA_CMSIS_CRL_THRESHOLDS','true')
```
To generate optimized code, thresholds are introduced . Forcing CMSIS function generation for all input lengths using this command can degrade performance.

Basic Math Operations

Operation	Wrappers Calling CMSIS Function	Supported Data Types	Input/Output Specifications	Parameter Specifications	Replaced Simulink^® Block
Element-wise real addition	`mw_arm_add_f64` `mw_arm_add_f32` `mw_arm_add_f16` `mw_arm_add_q31` `mw_arm_add_q15` `mw_arm_add_q7`	`double` `single` `half` `fixdt(true,32,)` `fixdt(true,16,)` `fixdt(true,8,*)`	Vector Matrix Both input and output must have the same word length. Both inputs must be of the same data type.	-	Sum
Real vector offset	`mw_arm_offset_f64` `mw_arm_offset_f32` `mw_arm_offset_f16` `mw_arm_offset_q31` `mw_arm_offset_q15` `mw_arm_offset_q7`			-	Sum
Element-wise real subtraction	`mw_arm_sub_f64` `mw_arm_sub_f32` `mw_arm_sub_f16` `mw_arm_sub_q31` `mw_arm_sub_q15` `mw_arm_sub_q7`	`double` `single` `half` `fixdt(true,32,)` `fixdt(true,16,)` `fixdt(true,8,*)`	Vector Matrix Both input and output must have the same word length. Both inputs must be of the same data type.	-	Sum
Element-wise real multiplication	`mw_arm_mult_f64` `mw_arm_mult_f32` `mw_arm_mult_f16` `mw_arm_mult_q31` `mw_arm_mult_q15` `mw_arm_mult_q7`	`double` `single` `half` `fixdt(true,32,)` `fixdt(true,16,)` `fixdt(true,8,*)`	Vector Matrix Both input and output must have the same word length. Ensure that inputs with floating-point data types, such as double, single, or half, are of the same type. For fixed-point inputs, both the inputs must be of equal word length. For inputs of type `fixdt(true,32,L1)` and `fixdt(true,32,L2)`, the output is of type `fixdt(true,32,L1+L2-31)` For inputs of type `fixdt(true,16,L1)` and `fixdt(true,16,L2)`, the output can be of type `fixdt(true,16,L1+L2-15)`. For inputs of type `fixdt(true,8,L1)` and `fixdt(true,8,L2)`, the output can be of type `fixdt(true,8,L1+L2-7)`.	-	Product, Matrix Multiply
Real vector scale	`mw_arm_scale_f64` `mw_arm_scale_f32` `mw_arm_scale_f16` `mw_arm_scale_q31` `mw_arm_scale_q15` `mw_arm_scale_q7`			Set the Multiplication parameter in the Gain block to `Matrix(Ku)`. The value of Gain* parameter must be a scalar.	Gain
Real absolute	`mw_arm_abs_f64` `mw_arm_abs_f32` `mw_arm_abs_f16` `mw_arm_abs_q31` `mw_arm_abs_q15` `mw_arm_abs_q7`	`double` `single` `half` `fixdt(true,32,31)` `fixdt(true,16,15)` `fixdt(true,8,7)`	Vector Matrix Both input and output must have the same word length.	-	Abs
Real dot product	`mw_arm_dot_prod_f64` `mw_arm_dot_prod_f32` `mw_arm_dot_prod_q31` `mw_arm_dot_prod_q15` `mw_arm_dot_prod_q7`	`double` `single` `fixdt(true,32,)` `fixdt(true,16,)` `fixdt(true,8,*)`	Vector Both input and output must have the same word length. Fixed-point real dot product: For inputs of type `fixdt(true,32,L1)` and `fixdt(true,32,L2)`, the output will be of type `fixdt(true,64,L1+L2-14)` For inputs of type `fixdt(true,16,L1)` and `fixdt(true,16,L2)`, the output can be of type `fixdt(true,64,L1+L2)`, `fixdt(true,32,L1+L2-6)`, or `fixdt(true,16,L1+L2-18)`.	-	Dot Product
Vector log real	`mw_arm_vlog_f64` `mw_arm_vlog_f32`	`double` `single`	Scalar Vector Matrix Both input and output must have the same word length.	Set the Function parameter of Math Function block to `log`.	Math Function
Vector exponential real	`mw_arm_vexp_f64` `mw_arm_vexp_f32` `mw_arm_vexp_f16`	`double` `single` `half`	Scalar Vector Matrix Both input and output must have the same word length.	Set the Function parameter of Math Function block to `exp`.	Math Function
Saturate	`mw_arm_clip_f32` `mw_arm_clip_q31` `mw_arm_clip_q15` `mw_arm_clip_q7`	`single` `fixdt(true,32,31)` `fixdt(true,16,15)` `fixdt(true,8,7)`	Vector Matrix Real inputs Both input and output must have the same word length. For the fixed-point data type, the lower limit and higher limit data type must match.	Upper and lower limits of the Saturation block must be scalar values. The fraction length of input, upper limit, and lower limit can be different.	Saturation

Complex Math Operations

Operation	Wrappers Calling CMSIS Function	Supported Data Types	Input/Output Specifications	Parameter Specifications	Replaced Simulink Block
Complex dot product	`mw_arm_cmplx_dot_prod_f32`	`single`	Vector and matrix Complex inputs	-	Dot Product
Complex-by-complex multiplication	`mw_arm_cmplx_mult_cmplx_f64` `mw_arm_cmplx_mult_cmplx_f32`	`double` `single`	Vector and matrix Both inputs must be complex	Set the Multiplication parameter in the Product, Matrix Multiply block to `Elementwise(.)`. The Output data type* parameter of Product, Matrix Multiply block must be set to the data type required.	Product, Matrix Multiply
Complex-by-real multiplication or real-by-complex multiplication	`mw_arm_cmplx_mult_real_f32` `mw_arm_real_mult_cmplx_f32`	`single`	Vector and matrix Real and complex inputs One input must be real and the other input must be complex.	Set the Multiplication parameter in the Product, Matrix Multiply block to `Elementwise(.)`. The Output data type* parameter of Product, Matrix Multiply block must be set to the data type required.	Product, Matrix Multiply

Matrix Operations

Operation	Wrappers Calling CMSIS Function	Supported Data Types	Input/Output Specifications	Parameter specifications	Replaced Simulink Block
Matrix multiplication	`mw_arm_mat_mult_f64` `mw_arm_mat_mult_f32` `mw_arm_mat_mult_q15` `mw_arm_mat_mult_q7`	`double` `single` `fixdt(true,16,)` `fixdt(true,8,)`	Matrix Real inputs	-	Product, Matrix Multiply
Complex matrix multiplication	`mw_arm_cmplx_mat_mult_f32` `mw_arm_cmplx_mat_mult_q15`	`single` `fixdt(true,16,*)`	Matrix Complex inputs	-	Product, Matrix Multiply
Matrix transpose	`mw_arm_mat_trans_f64` `mw_arm_mat_trans_f32` `mw_arm_mat_trans_f16` `mw_arm_mat_trans_q31` `mw_arm_mat_trans_q15` `mw_arm_mat_trans_q7`	`double` `single` `half` `fixdt(true,32,)` `fixdt(true,16,)` `fixdt(true,8,*)`	Matrix Real inputs	-	Transpose
Complex matrix transpose	`mw_arm_mat_cmplx_trans_f32` `mw_arm_mat_cmplx_trans_q31` `mw_arm_mat_cmplx_trans_q15`	`single` `fixdt(true,32,)` `fixdt(true,16,)`	Matrix Complex inputs	-	Transpose
Matrix-by-vector multiplication	`neon_mv_mul_u8x16`	`uint8`	Vector and matrix Number of rows and columns in first input must be a multiple of 16 to get code replacement. Number of rows in second input must be a multiple of 16 to get code replacement.	-	Product, Matrix Multiply
	`neon_mv_mul_u16x8`	`uint16`	Vector and matrix Number of rows and columns in first input must be a multiple of 8 to get code replacement. Number of rows in second input must be a multiple of 8 to get code replacement.
	`neon_mv_mul_u32x4`	`uint32`	Vector and matrix Number of rows and columns in first input must be a multiple of 4 to get code replacement. Number of rows in second input must be a multiple of 4 to get code replacement.
Matrix-by-matrix multiplication	`neon_mm_mul_u16x8`	`uint16`	Matrix Number of rows and columns in first and second input must be a multiple of 8 to get code replacement.
Matrix-by-matrix multiplication	`neon_mm_mul_u32x4`	`uint32`	Matrix Number of rows and columns in first and second input must be a multiple of 4 to get code replacement.

Supported Simulink Blocks with CMSIS Library for ARM Cortex-A Processors

Basic Math Operations

Complex Math Operations

Matrix Operations

See Also