Conditions for Code Replacement of MATLAB Functions with QHL

R2026b

Each MATLAB^® function that can be used with the Embedded Coder^® Support Package for Qualcomm^® Hexagon^® Processors requires specific conditions to allow code replacement by using the Qualcomm Hexagon library (QHL). You use this code replacement library to generate C/C++ code from a Simulink^® model or from MATLAB code.

When you satisfy the specified conditions, the generated C/C++ code uses QHL support to achieve optimized execution on Qualcomm Hexagon processors.

Basic Math Operations

QHL supports these MATLAB math functions for floating-point data type only when you configure the required properties, as specified in this table.

Function	Input Requirements	Property Requirements	Equivalent Functions
abs	Input must be a scalar, vector, or matrix Real values	Input length must be > 11	Vector, and matrix real inputs qhmath_abs_af
sqrt	Input must be a scalar, vector, or matrix Real values	-	Scalar real inputs qhmath_sqrt_f Vector, and matrix real inputs qhmath_sqrt_af
rsqrt (Fixed-Point Designer)	Input must be a scalar, vector, or matrix Real values	-	Scalar real inputs qhmath_rsqrt_f Vector, and matrix real inputs qhmath_rsqrt_af
log	Input must be a scalar, vector, or matrix Real and complex values	-	Scalar real inputs qhmath_log_f Vector, and matrix real inputs qhmath_log_af Scalar complex inputs mw_qhcomplex_clog_f Vector and matrix complex inputs mw_qhcomplex_clog_af
log2	Input must be a scalar, vector, or matrix Real values	-	Scalar real inputs qhmath_log2_f Vector, and matrix real inputs qhmath_log2_af
log10	Input must be a scalar, vector, or matrix Real values	-	Scalar real inputs qhmath_log10_f Vector, and matrix real inputs qhmath_log10_af
exp	Input must be a scalar, vector, or matrix Real and complex values	-	Scalar real inputs qhmath_cexp_f vector, and matrix real inputs qhmath_cexp_af scalar complex inputs mw_qhcomplex_cexp_f Vector, and Matrix complex inputs mw_qhcomplex_cexp_af
power	Input must be a scalar, vector, or matrix Real and complex values	-	Scalar, matrix real inputs (base) and vector, matrix real inputs (exponent) mw_qhmath_pow_f Vector real inputs (base) and scalar real inputs (exponent) qhmath_pow_af Scalar, vector, and matrix complex inputs (base and exponent) mw_qhcomplex_cpow_f
ceil	Input must be a scalar real value	-	Scalar real inputs qhmath_ceil
floor	Input must be a scalar real value	-	Scalar real inputs qhmath_floor
norm	Input must be a vector or matrix Real values	-	Vector and matrix real inputs qhblas_f_vector_norm_af
min	Input must be a vector or matrix Real values	Input length must be > 22	Vector and matrix real inputs mw_qhblas_f_vector_self_min_af
max	Input must be a vector or matrix Real values	Input length must be > 22	Vector and matrix real inputs mw_qhblas_f_vector_self_max_af

QHL supports these MATLAB math functions for fixed-point Q15 data type only when you configure the required properties, as specified in this table.

Function	Input Requirements	Input DataType	Property Requirements	Equivalent Functions
abs	Input must be a vector or matrix Real values	fixdt(1,16,15)	Input length must be > 8	Vector and matrix real inputs mw_qhmath_abs_ah
sqrt	Input must be a scalar, vector, or matrix Real unsigned values	fixdt(0,16,*)	-	Scalar real inputs qhmath_sqrt_h Vector and matrix real inputs qhmath_sqrt_ah
max	Input must be a vector or matrix Real values	fixdt(1,16,15)	Input length must be > 22	Vector and matrix real inputs mw_qhblas_h_vector_self_max_ah
min	Input must be a vector or matrix Real values	fixdt(1,16,15)	Input length must be > 22	Vector and matrix real inputs mw_qhblas_h_vector_self_min_ah

Trigonometric Operations

Function	Input Requirements	Property Requirements	Equivalent Functions
sin	Input must be a scalar, vector or matrix Real or complex Values	-	Scalar real inputs qhmath_sin_f Scalar, vector, and matrix real inputs qhmath_sin_af Scalar, vector, and matrix complex inputs mw_qhcomplex_csin_f
cos	Input must be a scalar, vector or matrix Real or complex Values	-	Scalar real inputs qhmath_cos_f Scalar, vector, and matrix real inputs qhmath_cos_af Scalar, vector, and matrix complex inputs mw_qhcomplex_ccos_f
tan	Input must be a scalar, vector, or matrix Real and complex values	-	Scalar real inputs qhmath_tan_f Vector and matrix real inputs qhmath_tan_af Scalar, vector, and matrix complex inputs mw_qhcomplex_ctan_f
asin	Input must be a scalar, vector or matrix Real and complex values	-	Scalar real inputs qhmath_asin_f Vector, and matrix real inputs qhmath_asin_af Scalar complex inputs mw_qhcomplex_casin_f Vector and matrix complex inputs mw_qhcomplex_casin_af
acos	Input must be a scalar, vector or matrix Real and complex values	-	Scalar real inputs qhmath_acos_f Vector, and matrix real inputs qhmath_acos_af Scalar complex inputs mw_qhcomplex_cacos_f Vector and matrix complex inputs mw_qhcomplex_cacos_af
atan	Input must be a scalar, vector or matrix Real and complex values	-	Scalar real inputs qhmath_atan_f Vector and matrix real inputs qhmath_atan_af Scalar complex inputs mw_qhcomplex_catan_f Vector and matrix complex inputs mw_qhcomplex_catan_af
sinh	Input must be a scalar, vector or matrix Real and complex values	-	Scalar real inputs qhmath_asinh_f Vector and matrix real inputs qhmath_asinh_af Scalar complex inputs mw_qhcomplex_csinh_f Vector and matrix complex inputs mw_qhcomplex_csinh_af
cosh	Input must be a scalar, vector or matrix Real and complex values	-	Scalar real inputs qhmath_acosh_f Vector and matrix real inputs qhmath_acosh_af Scalar complex inputs mw_qhcomplex_ccosh_f Vector and matrix complex inputs mw_qhcomplex_ccosh_af
tanh	Input must be a scalar, vector, or matrix Real and complex values	-	Scalar real inputs qhmath_ctanh_f Vector and matrix real inputs qhmath_ctanh_af Scalar complex inputs mw_qhcomplex_ctanh_f Vector and matrix complex inputs mw_qhcomplex_ccosh_af
hypot	Input must be a scalar real value	-	Scalar real inputs qhmath_hypot

Fourier Transform Operations

Function	Input Requirements	Output Requirements	Input Alignment	Output Alignment	Property Requirements	Equivalent Functions
fft	Input must be real Supports input sizes that are powers of 2, starting from 16.	Output must be complex	8	8	Note To enable code replacement, use the syntax fft(x). The code generator replaces only this form with a QHL implementation. If you specify additional inputs such as n or dim (for example, fft(X,n) or fft(X,n,dim)), the code generator uses the default generated code.	mw_qhdsp_acf_r1dfft_af_opt_builtin
fft	Input must be complex Supports input sizes that are powers of 2, starting from 8.	Output must be complex	size*8	8		mw_qhdsp_c1dfft_acf_builtin
ifft	Input must be complex Supports input sizes that are powers of 2, starting from 16.	Output must be complex	size*8	8	Note To enable code replacement, use the syntax ifft(x). The code generator replaces only this form with a QHL implementation. If you specify additional inputs such as n or dim (for example, ifft(X,n) or ifft(X,n,dim)), the code generator uses the default generated code.	mw_qhdsp_c1difft_acf_builtin