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

Embedded Coder^® Support Package for ARM^® Cortex^®-M Processors provides a code replacement library (CRL) for common microcontroller software interface standard (CMSIS) functions.

CMSIS CRL supports these processors:

• Cortex-M0

• Cortex-M0+

• Cortex-M3

• Cortex-M4

• Cortex-M7

• Cortex-M33

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

Note

• To get the code replacement, in the code generation configuration settings, set CodeReplacementLibrary to ARM Cortex-M.

• Select the Saturate on integer overflow option in Simulink^® blocks for CRL replacements. Refer to Enable Saturate on Integer Overflow on a Block section in Collect Saturation on Integer Overflow Coverage (Simulink Coverage).

• If you encounter situations where the ARM Cortex-M CMSIS CRL does not provide code replacements due to the input vector length being below the threshold, you can enable CMSIS function generation for all input lengths. To do this, set the environment variable DISABLE_ARMCORTEXM_CMSIS_CRL_THRESHOLDS to true using the following command:

  setenv('DISABLE_ARMCORTEXM_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

Basic Math Operations

Operation	Wrappers calling CMSIS function	Supported data types	Input/Output specifications	Parameter specifications	Replaced Simulink block
Absolute	mw_arm_abs_f32	single	Vector and matrix Real inputs	-	Abs
	mw_arm_abs_q31	fixdt(true,32,31)	Vector and matrix Real inputs
	mw_arm_abs_q15	fixdt(true,16,15)	Vector and matrix Real inputs
	mw_arm_abs_q7	fixdt(true,8,7)	Vector and matrix Real inputs
Addition	mw_arm_add_f32	single	Vector and matrix Real and complex inputs Note Using a vector input and a vector bias value in the Bias block will also trigger code replacement.	-	Add
	mw_arm_add_q31	fixdt(true,32,*)	Vector and matrix Real and complex inputs Fraction length of inputs must be same. Note For real inputs, the CRL generates code replacement when the word length is 32, even if the fraction length of the inputs differ.
	mw_arm_add_q15	fixdt(true,16,*)	Vector and matrix Real and complex inputs Fraction length of inputs must be same. Note For real inputs, the CRL generates code replacement when the word length is 16, even if the fraction length of the inputs differ.
	mw_arm_add_q7	fixdt(true,8,*)	Vector and matrix Real and complex inputs Fraction length of inputs must be same. Note For real inputs, the CRL generates code replacement when the word length is 8, even if the fraction length of the inputs differ.
Bias/Offset	mw_arm_bias_1_f32	single	Scalar input Real inputs	Bias parameter in the Bias block can be vector and matrix.	Bias
	mw_arm_bias_1_q31	fixdt(true,32,*)	Scalar input Real inputs Supports different input and bias word lengths. Supports different input and bias fraction length
	mw_arm_bias_1_q15	fixdt(true,16,*)
	mw_arm_bias_1_q7	fixdt(true,8,*)
	mw_arm_bias_2_f32	single	Vector and matrix Real inputs	Bias parameter in the Bias block must be a scalar.
	mw_arm_bias_2_q31	fixdt(true,32,*)	Vector and matrix Real inputs Supports different input and bias word lengths. Supports different input and bias fraction length
	mw_arm_bias_2_q15	fixdt(true,16,*)
	mw_arm_bias_2_q7	fixdt(true,8,*)
Dot product	mw_arm_dot_prod_f32	single	Vector Real inputs Fraction length of output must be equal to the sum of input fraction lengths.	-	Dot Product
	mw_arm_dot_prod_q31	fixdt(true,32,*)	Vector Real inputs Both inputs must have same word length Note Set the output word length to 64 for an input word length of 32. Fraction length of output msut be equal to the sum of input fraction lengths.
	mw_arm_dot_prod_q15	fixdt(true,16,*)	Vector Real inputs Both inputs must have same word length Note Set the output word length to 64 for an input word length of 16. Fraction length of output must be equal to the sum of input fraction lengths.
	mw_arm_dot_prod_q7	fixdt(true,8,*)	Vector Real inputs Both inputs must have same word length Note Set the output word length to 32 for an input word length of 8. Fraction length of output must be equal to the sum of input fraction lengths.
Exponential	mw_arm_vexp_f32	single	Vector and matrix Real inputs	Set the Function parameter of Math Function block to exp.	Math Function
Logarithm	mw_arm_vlog_f32	single	Vector and matrix Real inputs	Set the Function parameter of Math Function to log.	Math Function
Multiplication	mw_arm_mult_f32	single	Vector and matrix Real inputs Note For multiplication operation, one input can be a vector while the other input can be a matrix.	-	Product, Matrix Multiply
	mw_arm_mult_q15	fixdt(true,16,*)	Vector and matrix Real inputs Note For multiplication operation, one input can be a vector while the other input can be a matrix. Sum of input fraction length must be greater than 15. Fraction length of output must be 15 less than sum of input fraction length.
	mw_arm_mult_q7	fixdt(true,8,*)	Vector and matrix Real inputs Note For multiplication operation, one input can be a vector while the other input can be a matrix. Sum of input fraction length must be greater than 7. Fraction length of output must be 7 less than sum of input fraction length.
Saturate/Limit/Clipping	mw_arm_saturate_f32	single	Vector and matrix Real inputs	Upper and lower limits of the Saturation block must be scalar values	Saturation
	mw_arm_saturate_q31	fixdt(true,32,*)	Vector and matrix Real inputs	Upper and lower limits of the Saturation block must be scalar values. The word length of input, upper limit, and lower limit must be 32. The fraction length of input, upper limit, and lower limit can be different.
	mw_arm_saturate_q15	fixdt(true,16,*)	Vector and matrix Real inputs	Upper and lower limits of the Saturation block must be scalar values. The word length of input, upper limit, and lower limit must be 16. The fraction length of input, upper limit, and lower limit can be different.
	mw_arm_saturate_q7	fixdt(true,8,*)	Vector and matrix Real inputs	Upper and lower limits of the Saturation block must be scalar values. The word length of input, upper limit, and lower limit must be 8. The fraction length of input, upper limit, and lower limit can be different.
Scale/Gain	mw_arm_scale_1_f32	single	Vector and matrix Real inputs	Set the Multiplication parameter in the Gain block to Matrix(K*u). The Gain parameter value of the Gain must be a scalar	Gain
	mw_arm_scale_1_q15	fixdt(true,16,*)	Vector and matrix Real inputs Supports different input, scalar gain value, and word lengths.	The fraction length of output must be less than the sum of fraction length of inputs. Set the Multiplication parameter in the Gain block to Matrix(K*u). The Gain parameter value of the Gain block must be a scalar
	mw_arm_scale_1_q7	fixdt(true,8,*)
	mw_arm_scale_2_f32	single	Vector and matrix Real inputs One input can be a vector or matrix, while the other input to the Product, Matrix Multiply block must be a scalar.	Set the Multiplication parameter in the Product, Matrix Multiply block to Element-wise(.*).
	mw_arm_scale_2_q15	fixdt(true,16,*)
	mw_arm_scale_2_q7	fixdt(true,8,*)
Subtraction	mw_arm_sub_f32	single	Vector and matrix Real and complex inputs	-	Sub
	mw_arm_sub_q31	fixdt(true,32,*)	Vector and matrix Real and complex inputs Fraction length of inputs must be same. Note For real inputs, the CRL generates code replacement when the word length is 32, even if the fraction lengths of the inputs differ.
	mw_arm_sub_q15	fixdt(true,16,*)	Vector and matrix Real and complex inputs Fraction length of inputs must be same. Note For real inputs, the CRL generates code replacement when the word length is 16, even if the fraction lengths of the inputs differ.
	mw_arm_sub_q7	fixdt(true,8,*)	Vector and matrix Real and complex inputs Fraction length of inputs must be same. Note For real inputs, the CRL generates code replacement when the word length is 8, even if the fraction lengths of the inputs differ.
Square root	mw_arm_sqrt_f32	single	Vector and matrix Real inputs	-	Sqrt
Unary minus/Negate	mw_arm_uminus_f32	single	Vector and matrix Real inputs	-	Unary Minus
	mw_arm_uminus_q31	fixdt(true,32,*)
	mw_arm_uminus_q15	fixdt(true,16,*)
	mw_arm_uminus_q7	fixdt(true,8,*)

Shift Operations

Shift Operations

Operation	Wrappers calling CMSIS function	Supported data types	Input/Output specifications	Replaced Simulink block
Right shift	mw_arm_rshift_q31	fixdt(true,32,31)	Vector and matrix Real inputs	Shift Arithmetic
	mw_arm_rshift_q15	fixdt(true,16,15)	Vector and matrix Real inputs
	mw_arm_rshift_q7	fixdt(true,8,7)	Vector and matrix Real inputs

Data Cast Operations

Data Cast Operations

Operation	Wrappers calling CMSIS function	Supported data types	Input/Output specifications	Parameter specifications	Replaced Simulink block
Cast float to fixed point	mw_arm_float_to_q15	single	Vector and matrix Real inputs	The Output data type parameter of Data Type Conversion block must be set to fixdt(1,16,15)	Data Type Conversion
	mw_arm_float_to_q7	single	Vector and matrix Real inputs	The Output data type parameter of Data Type Conversion block must be set to fixdt(1,8,7)
Cast q31 to float	mw_arm_q31_to_float	fixdt(true,32,31)	Vector and matrix Real inputs	The Output data type parameter of Data Type Conversion block must be set to single
Cast q31 to q15	mw_arm_q31_to_q15	fixdt(true,32,31)	Vector and matrix Real inputs	The Output data type parameter of Data Type Conversion block must be set to fixdt(1,16,15)
Cast q31 to q7	mw_arm_q31_to_q7	fixdt(true,32,31)	Vector and matrix Real inputs	The Output data type parameter of Data Type Conversion block must be set to fixdt(1,8,7)
Cast q15 to float	mw_arm_q15_to_float	fixdt(true,16,15)	Vector and matrix Real inputs	The Output data type parameter of Data Type Conversion block must be set to single
Cast q15 to q31	mw_arm_q15_to_q31	fixdt(true,16,15)	Vector and matrix Real inputs	The Output data type parameter of Data Type Conversion block must be set to fixdt(1,32,31)
Cast q15 to q7	mw_arm_q15_to_q7	fixdt(true,16,15)	Vector and matrix Real inputs	The Output data type parameter of Data Type Conversion block must be set to fixdt(1,8,7)
Cast q7 to float	mw_arm_q7_to_float	fixdt(true,8,7)	Vector and matrix Real inputs	The Output data type parameter of Data Type Conversion block must be set to single
Cast q7 to q31	mw_arm_q7_to_q31	fixdt(true,8,7)	Vector and matrix Real inputs	The Output data type parameter of Data Type Conversion block must be set to fixdt(1,32,31)
Cast q7 to q15	mw_arm_q7_to_q15	fixdt(true,8,7)	Vector and matrix Real inputs	The Output data type parameter of Data Type Conversion block must be set to fixdt(1,16,15)

Trigonometric Operations

Trigonometric Operations

Operation	Wrappers calling CMSIS function	Supported data types	Input/Output specifications	Parameter specifications	Replaced Simulink block
sine	mw_arm_sin_f32	single	Vector and matrix Real inputs	The Approximation method parameter of Trigonometric Function block must be set to Lookup Look-up table errors cause a numerical mismatch on the order of 5e-03.	Trigonometric Function
cosine	mw_arm_cos_f32	single

Complex Math Operations

Complex Math Operations

Operation	Wrappers calling CMSIS function	Supported data types	Input/Output specifications	Parameter specifications	Replaced Simulink block
Complex conjugate	mw_arm_cmplx_conj_f32	single	Vector and matrix Complex inputs	Set the Function parameter of Math Function to conj.	Math Function
	mw_arm_cmplx_conj_q31	fixdt(true,32,31)
	mw_arm_cmplx_conj_q15	fixdt(true,16,15)
Complex-by-complex multiplication	mw_arm_cmplx_mult_cmplx_f32	single	Vector and matrix Both inputs must be complex	Set the Multiplication parameter of Math Function in the Product, Matrix Multiply block to Elementwise(.*).	Product, Matrix Multiply
	mw_arm_cmplx_mult_cmplx_q31	fixdt(true,32,31)	Vector and matrix Both inputs must be complex	The Output data type parameter of Product, Matrix Multiply block must be set to fixdt(1,32,29) Set the Multiplication parameter in the Product, Matrix Multiply block to Elementwise(.*).
	mw_arm_cmplx_mult_cmplx_q15	fixdt(true,16,15)	Vector and matrix Both inputs must be complex	The Output data type parameter of Product, Matrix Multiply block must be set to fixdt(1,16,13) Set the Multiplication parameter in the Product, Matrix Multiply block to Elementwise(.*).
Complex-by-real multiplication	mw_arm_cmplx_mult_real_f32	single	Vector and matrix Real and complex inputs One input should be complex, and the other input should be real.	Set the Multiplication parameter in the Product, Matrix Multiply block to Elementwise(.*).	Product, Matrix Multiply
	mw_arm_cmplx_mult_real_q31	fixdt(true,32,31)	Vector and matrix Real and complex inputs One input should be complex, and the other input should be real.	The Output data type parameter of Product, Matrix Multiply block must be set to fixdt(1,32,31) Set the Multiplication parameter in the Product, Matrix Multiply block to Elementwise(.*).
	mw_arm_cmplx_mult_real_q15	fixdt(true,16,15)	Vector and matrix Real and complex inputs One input should be complex, and the other input should be real.	The Output data type parameter of Product, Matrix Multiply block must be set to fixdt(1,16,15) Set the Multiplication parameter in the Product, Matrix Multiply block to Elementwise(.*).

Logical Operations

Logical Operations

Operation	Wrappers calling CMSIS function	Supported data types	Input/Output specifications	Parameter specifications	Replaced Simulink block
Logical AND	mw_arm_and_32	uint32 fixdt(false,32,*)	Vector and matrix Both inputs must be uint32 or fixdt(false,32,*). Both inputs must have same fraction length.	Set the Operator parameter in the Bitwise Operator block to AND.	AND
	mw_arm_and_uint16	uint16 fixdt(false,16,*)	Vector and matrix Both inputs must be uint16 or fixdt(false,16,*). Both inputs must have same fraction length.
	mw_arm_and_uint8	uint8 fixdt(false,8,*)	Vector and matrix Both inputs must be uint8 or fixdt(false,8,*). Both inputs must have same fraction length.
Logical OR	mw_arm_or_uint32	uint32 fixdt(false,32,*)	Vector and matrix Both inputs must be uint32 or fixdt(false,32,*). Both inputs must have same fraction length.	Set the Operator parameter in the Bitwise Operator block to OR.	OR
	mw_arm_or_uint16	uint16 fixdt(false,16,*)	Vector and matrix Both inputs must be uint16 or fixdt(false,16,*). Both inputs must have same fraction length.
	mw_arm_or_uint8	uint8 fixdt(false,8,*)	Vector and matrix Both inputs must be uint8 or fixdt(false,8,*). Both inputs must have same fraction length.
Logical NOT	mw_arm_not_uint32	uint32 fixdt(false,32,*)	Vector and matrix	Set the Operator parameter in the Bitwise Operator block to NOT.	NOT
	mw_arm_not_uint16	uint16 fixdt(false,16,*)	Vector and matrix
	mw_arm_not_uint8	uint8 fixdt(false,8,*)	Vector and matrix
Logical XOR	mw_arm_xor_uint32	uint32 fixdt(false,32,*)	Vector and matrix Both inputs must be uint32 or fixdt(false,32,*). Both inputs must have same fraction length.	Set the Operator parameter in the Bitwise Operator block to XOR.	XOR
	mw_arm_xor_uint16	uint16 fixdt(false,16,*)	Vector and matrix Both inputs must be uint16 or fixdt(false,16,*). Both inputs must have same fraction length.
	mw_arm_xor_uint8	uint8 fixdt(false,8,*)	Vector and matrix Both inputs must be uint8 or fixdt(false,8,*). Both inputs must have same fraction length.

Matrix Operations

Matrix Operations

Operation	Wrappers calling CMSIS function	Supported data types	Input/Output specifications	Parameter specifications	Replaced Simulink block
Matrix multiplication	mw_arm_mat_mult_f32	single	Vector and matrix	Set the Multiplication parameter of Product, Matrix Multiply block to Matrix(*).	Product, Matrix Multiply
	mw_arm_mat_mult_q15	fixdt(1,16,*)	Vector and matrix Both inputs must be fixdt(1,16,*). Note The output with CRL replacement differs from the Simulink output because the CMSIS APIs use a higher accumulator, whereas Simulink uses an accumulator with a size equal to the input word length	Set the Multiplication parameter of Product, Matrix Multiply block to Matrix(*). The Output data type parameter of Product, Matrix Multiply block must be set to fixdt(1,16,*).
	mw_arm_mat_mult_q7	fixdt(1,8,*)	Vector and matrix Both inputs must be fixdt(1,8,*). Note The output with CRL replacement differs from the Simulink output because the CMSIS APIs use a higher accumulator, whereas Simulink uses an accumulator with a size equal to the input word length	Set the Multiplication parameter of Product, Matrix Multiply block to Matrix(*). The Output data type parameter of Product, Matrix Multiply block must be set to fixdt(1,8,*).
Complex matrix multiplication	mw_arm_mat_cmplx_mult_f32	single	Vector and matrix Complex values Both inputs must be of same word length.	Set the Multiplication parameter of Product, Matrix Multiply block to Matrix(*).	Product, Matrix Multiply
	mw_arm_mat_cmplx_mult_q15	fixdt(true,16,*)		Set the Multiplication parameter of Product, Matrix Multiply block to Matrix(*). The Output data type parameter of Product, Matrix Multiply block must be set to fixdt(1,16,*).
Matrix transpose	mw_arm_trans_f32	single	Matrix Real inputs	-	Transpose
	mw_arm_trans_q31	fixdt(true,32,*)
	mw_arm_trans_q15	fixdt(true,16,*)
	mw_arm_trans_q7	fixdt(true,8,*)
	mw_arm_mat_cmplx_trans_f32	single	Matrix Complex inputs
	mw_arm_mat_cmplx_trans_q31	fixdt(true,32,*)
	mw_arm_mat_cmplx_trans_q15	fixdt(true,16,*)