arb_poly.h – polynomials over the real numbers

An arb_poly_t represents a polynomial over the real numbers, implemented as an array of coefficients of type arb_struct.

Most functions are provided in two versions: an underscore method which operates directly on pre-allocated arrays of coefficients and generally has some restrictions (such as requiring the lengths to be nonzero and not supporting aliasing of the input and output arrays), and a non-underscore method which performs automatic memory management and handles degenerate cases.

Types, macros and constants

arb_poly_struct
arb_poly_t

Contains a pointer to an array of coefficients (coeffs), the used length (length), and the allocated size of the array (alloc).

An arb_poly_t is defined as an array of length one of type arb_poly_struct, permitting an arb_poly_t to be passed by reference.

Memory management

void arb_poly_init(arb_poly_t poly)

Initializes the polynomial for use, setting it to the zero polynomial.

void arb_poly_clear(arb_poly_t poly)

Clears the polynomial, deallocating all coefficients and the coefficient array.

void arb_poly_fit_length(arb_poly_t poly, slong len)

Makes sure that the coefficient array of the polynomial contains at least len initialized coefficients.

void _arb_poly_set_length(arb_poly_t poly, slong len)

Directly changes the length of the polynomial, without allocating or deallocating coefficients. The value should not exceed the allocation length.

void _arb_poly_normalise(arb_poly_t poly)

Strips any trailing coefficients which are identical to zero.

slong arb_poly_allocated_bytes(const arb_poly_t x)

Returns the total number of bytes heap-allocated internally by this object. The count excludes the size of the structure itself. Add sizeof(arb_poly_struct) to get the size of the object as a whole.

Basic manipulation

slong arb_poly_length(const arb_poly_t poly)

Returns the length of poly, i.e. zero if poly is identically zero, and otherwise one more than the index of the highest term that is not identically zero.

slong arb_poly_degree(const arb_poly_t poly)

Returns the degree of poly, defined as one less than its length. Note that if one or several leading coefficients are balls containing zero, this value can be larger than the true degree of the exact polynomial represented by poly, so the return value of this function is effectively an upper bound.

int arb_poly_is_zero(const arb_poly_t poly)
int arb_poly_is_one(const arb_poly_t poly)
int arb_poly_is_x(const arb_poly_t poly)

Returns 1 if poly is exactly the polynomial 0, 1 or x respectively. Returns 0 otherwise.

void arb_poly_zero(arb_poly_t poly)
void arb_poly_one(arb_poly_t poly)

Sets poly to the constant 0 respectively 1.

void arb_poly_set(arb_poly_t dest, const arb_poly_t src)

Sets dest to a copy of src.

void arb_poly_set_round(arb_poly_t dest, const arb_poly_t src, slong prec)

Sets dest to a copy of src, rounded to prec bits.

void arb_poly_set_trunc(arb_poly_t dest, const arb_poly_t src, slong n)
void arb_poly_set_trunc_round(arb_poly_t dest, const arb_poly_t src, slong n, slong prec)

Sets dest to a copy of src, truncated to length n and rounded to prec bits.

void arb_poly_set_coeff_si(arb_poly_t poly, slong n, slong c)
void arb_poly_set_coeff_arb(arb_poly_t poly, slong n, const arb_t c)

Sets the coefficient with index n in poly to the value c. We require that n is nonnegative.

void arb_poly_get_coeff_arb(arb_t v, const arb_poly_t poly, slong n)

Sets v to the value of the coefficient with index n in poly. We require that n is nonnegative.

arb_poly_get_coeff_ptr(poly, n)

Given \(n \ge 0\), returns a pointer to coefficient n of poly, or NULL if n exceeds the length of poly.

void _arb_poly_shift_right(arb_ptr res, arb_srcptr poly, slong len, slong n)
void arb_poly_shift_right(arb_poly_t res, const arb_poly_t poly, slong n)

Sets res to poly divided by \(x^n\), throwing away the lower coefficients. We require that n is nonnegative.

void _arb_poly_shift_left(arb_ptr res, arb_srcptr poly, slong len, slong n)
void arb_poly_shift_left(arb_poly_t res, const arb_poly_t poly, slong n)

Sets res to poly multiplied by \(x^n\). We require that n is nonnegative.

void arb_poly_truncate(arb_poly_t poly, slong n)

Truncates poly to have length at most n, i.e. degree strictly smaller than n. We require that n is nonnegative.

slong arb_poly_valuation(const arb_poly_t poly)

Returns the degree of the lowest term that is not exactly zero in poly. Returns -1 if poly is the zero polynomial.

Conversions

void arb_poly_set_fmpz_poly(arb_poly_t poly, const fmpz_poly_t src, slong prec)
void arb_poly_set_fmpq_poly(arb_poly_t poly, const fmpq_poly_t src, slong prec)
void arb_poly_set_si(arb_poly_t poly, slong src)

Sets poly to src, rounding the coefficients to prec bits.

Input and output

void arb_poly_printd(const arb_poly_t poly, slong digits)

Prints the polynomial as an array of coefficients, printing each coefficient using arb_printd.

void arb_poly_fprintd(FILE * file, const arb_poly_t poly, slong digits)

Prints the polynomial as an array of coefficients to the stream file, printing each coefficient using arb_fprintd.

Random generation

void arb_poly_randtest(arb_poly_t poly, flint_rand_t state, slong len, slong prec, slong mag_bits)

Creates a random polynomial with length at most len.

Comparisons

int arb_poly_contains(const arb_poly_t poly1, const arb_poly_t poly2)
int arb_poly_contains_fmpz_poly(const arb_poly_t poly1, const fmpz_poly_t poly2)
int arb_poly_contains_fmpq_poly(const arb_poly_t poly1, const fmpq_poly_t poly2)

Returns nonzero iff poly1 contains poly2.

int arb_poly_equal(const arb_poly_t A, const arb_poly_t B)

Returns nonzero iff A and B are equal as polynomial balls, i.e. all coefficients have equal midpoint and radius.

int _arb_poly_overlaps(arb_srcptr poly1, slong len1, arb_srcptr poly2, slong len2)
int arb_poly_overlaps(const arb_poly_t poly1, const arb_poly_t poly2)

Returns nonzero iff poly1 overlaps with poly2. The underscore function requires that len1 ist at least as large as len2.

int arb_poly_get_unique_fmpz_poly(fmpz_poly_t z, const arb_poly_t x)

If x contains a unique integer polynomial, sets z to that value and returns nonzero. Otherwise (if x represents no integers or more than one integer), returns zero, possibly partially modifying z.

Bounds

void _arb_poly_majorant(arb_ptr res, arb_srcptr poly, slong len, slong prec)
void arb_poly_majorant(arb_poly_t res, const arb_poly_t poly, slong prec)

Sets res to an exact real polynomial whose coefficients are upper bounds for the absolute values of the coefficients in poly, rounded to prec bits.

Arithmetic

void _arb_poly_add(arb_ptr C, arb_srcptr A, slong lenA, arb_srcptr B, slong lenB, slong prec)

Sets {C, max(lenA, lenB)} to the sum of {A, lenA} and {B, lenB}. Allows aliasing of the input and output operands.

void arb_poly_add(arb_poly_t C, const arb_poly_t A, const arb_poly_t B, slong prec)
void arb_poly_add_si(arb_poly_t C, const arb_poly_t A, slong B, slong prec)

Sets C to the sum of A and B.

void _arb_poly_sub(arb_ptr C, arb_srcptr A, slong lenA, arb_srcptr B, slong lenB, slong prec)

Sets {C, max(lenA, lenB)} to the difference of {A, lenA} and {B, lenB}. Allows aliasing of the input and output operands.

void arb_poly_sub(arb_poly_t C, const arb_poly_t A, const arb_poly_t B, slong prec)

Sets C to the difference of A and B.

void arb_poly_add_series(arb_poly_t C, const arb_poly_t A, const arb_poly_t B, slong len, slong prec)

Sets C to the sum of A and B, truncated to length len.

void arb_poly_sub_series(arb_poly_t C, const arb_poly_t A, const arb_poly_t B, slong len, slong prec)

Sets C to the difference of A and B, truncated to length len.

void arb_poly_neg(arb_poly_t C, const arb_poly_t A)

Sets C to the negation of A.

void arb_poly_scalar_mul_2exp_si(arb_poly_t C, const arb_poly_t A, slong c)

Sets C to A multiplied by \(2^c\).

void arb_poly_scalar_mul(arb_poly_t C, const arb_poly_t A, const arb_t c, slong prec)

Sets C to A multiplied by c.

void arb_poly_scalar_div(arb_poly_t C, const arb_poly_t A, const arb_t c, slong prec)

Sets C to A divided by c.

void _arb_poly_mullow_classical(arb_ptr C, arb_srcptr A, slong lenA, arb_srcptr B, slong lenB, slong n, slong prec)
void _arb_poly_mullow_block(arb_ptr C, arb_srcptr A, slong lenA, arb_srcptr B, slong lenB, slong n, slong prec)
void _arb_poly_mullow(arb_ptr C, arb_srcptr A, slong lenA, arb_srcptr B, slong lenB, slong n, slong prec)

Sets {C, n} to the product of {A, lenA} and {B, lenB}, truncated to length n. The output is not allowed to be aliased with either of the inputs. We require \(\mathrm{lenA} \ge \mathrm{lenB} > 0\), \(n > 0\), \(\mathrm{lenA} + \mathrm{lenB} - 1 \ge n\).

The classical version uses a plain loop. This has good numerical stability but gets slow for large n.

The block version decomposes the product into several subproducts which are computed exactly over the integers.

It first attempts to find an integer \(c\) such that \(A(2^c x)\) and \(B(2^c x)\) have slowly varying coefficients, to reduce the number of blocks.

The scaling factor \(c\) is chosen in a quick, heuristic way by picking the first and last nonzero terms in each polynomial. If the indices in \(A\) are \(a_2, a_1\) and the log-2 magnitudes are \(e_2, e_1\), and the indices in \(B\) are \(b_2, b_1\) with corresponding magnitudes \(f_2, f_1\), then we compute \(c\) as the weighted arithmetic mean of the slopes, rounded to the nearest integer:

\[c = \left\lfloor \frac{(e_2 - e_1) + (f_2 + f_1)}{(a_2 - a_1) + (b_2 - b_1)} + \frac{1}{2} \right \rfloor.\]

This strategy is used because it is simple. It is not optimal in all cases, but will typically give good performance when multiplying two power series with a similar decay rate.

The default algorithm chooses the classical algorithm for short polynomials and the block algorithm for long polynomials.

If the input pointers are identical (and the lengths are the same), they are assumed to represent the same polynomial, and its square is computed.

void arb_poly_mullow_classical(arb_poly_t C, const arb_poly_t A, const arb_poly_t B, slong n, slong prec)
void arb_poly_mullow_ztrunc(arb_poly_t C, const arb_poly_t A, const arb_poly_t B, slong n, slong prec)
void arb_poly_mullow_block(arb_poly_t C, const arb_poly_t A, const arb_poly_t B, slong n, slong prec)
void arb_poly_mullow(arb_poly_t C, const arb_poly_t A, const arb_poly_t B, slong n, slong prec)

Sets C to the product of A and B, truncated to length n. If the same variable is passed for A and B, sets C to the square of A truncated to length n.

void _arb_poly_mul(arb_ptr C, arb_srcptr A, slong lenA, arb_srcptr B, slong lenB, slong prec)

Sets {C, lenA + lenB - 1} to the product of {A, lenA} and {B, lenB}. The output is not allowed to be aliased with either of the inputs. We require \(\mathrm{lenA} \ge \mathrm{lenB} > 0\). This function is implemented as a simple wrapper for _arb_poly_mullow().

If the input pointers are identical (and the lengths are the same), they are assumed to represent the same polynomial, and its square is computed.

void arb_poly_mul(arb_poly_t C, const arb_poly_t A, const arb_poly_t B, slong prec)

Sets C to the product of A and B. If the same variable is passed for A and B, sets C to the square of A.

void _arb_poly_inv_series(arb_ptr Q, arb_srcptr A, slong Alen, slong len, slong prec)

Sets {Q, len} to the power series inverse of {A, Alen}. Uses Newton iteration.

void arb_poly_inv_series(arb_poly_t Q, const arb_poly_t A, slong n, slong prec)

Sets Q to the power series inverse of A, truncated to length n.

void _arb_poly_div_series(arb_ptr Q, arb_srcptr A, slong Alen, arb_srcptr B, slong Blen, slong n, slong prec)

Sets {Q, n} to the power series quotient of {A, Alen} by {B, Blen}. Uses Newton iteration followed by multiplication.

void arb_poly_div_series(arb_poly_t Q, const arb_poly_t A, const arb_poly_t B, slong n, slong prec)

Sets Q to the power series quotient A divided by B, truncated to length n.

void _arb_poly_div(arb_ptr Q, arb_srcptr A, slong lenA, arb_srcptr B, slong lenB, slong prec)
void _arb_poly_rem(arb_ptr R, arb_srcptr A, slong lenA, arb_srcptr B, slong lenB, slong prec)
void _arb_poly_divrem(arb_ptr Q, arb_ptr R, arb_srcptr A, slong lenA, arb_srcptr B, slong lenB, slong prec)
int arb_poly_divrem(arb_poly_t Q, arb_poly_t R, const arb_poly_t A, const arb_poly_t B, slong prec)

Performs polynomial division with remainder, computing a quotient \(Q\) and a remainder \(R\) such that \(A = BQ + R\). The implementation reverses the inputs and performs power series division.

If the leading coefficient of \(B\) contains zero (or if \(B\) is identically zero), returns 0 indicating failure without modifying the outputs. Otherwise returns nonzero.

void _arb_poly_div_root(arb_ptr Q, arb_t R, arb_srcptr A, slong len, const arb_t c, slong prec)

Divides \(A\) by the polynomial \(x - c\), computing the quotient \(Q\) as well as the remainder \(R = f(c)\).

Composition

void _arb_poly_taylor_shift_horner(arb_ptr g, const arb_t c, slong n, slong prec)
void arb_poly_taylor_shift_horner(arb_poly_t g, const arb_poly_t f, const arb_t c, slong prec)
void _arb_poly_taylor_shift_divconquer(arb_ptr g, const arb_t c, slong n, slong prec)
void arb_poly_taylor_shift_divconquer(arb_poly_t g, const arb_poly_t f, const arb_t c, slong prec)
void _arb_poly_taylor_shift_convolution(arb_ptr g, const arb_t c, slong n, slong prec)
void arb_poly_taylor_shift_convolution(arb_poly_t g, const arb_poly_t f, const arb_t c, slong prec)
void _arb_poly_taylor_shift(arb_ptr g, const arb_t c, slong n, slong prec)
void arb_poly_taylor_shift(arb_poly_t g, const arb_poly_t f, const arb_t c, slong prec)

Sets g to the Taylor shift \(f(x+c)\), computed respectively using an optimized form of Horner’s rule, divide-and-conquer, a single convolution, and an automatic choice between the three algorithms.

The underscore methods act in-place on g = f which has length n.

void _arb_poly_compose_horner(arb_ptr res, arb_srcptr poly1, slong len1, arb_srcptr poly2, slong len2, slong prec)
void arb_poly_compose_horner(arb_poly_t res, const arb_poly_t poly1, const arb_poly_t poly2, slong prec)
void _arb_poly_compose_divconquer(arb_ptr res, arb_srcptr poly1, slong len1, arb_srcptr poly2, slong len2, slong prec)
void arb_poly_compose_divconquer(arb_poly_t res, const arb_poly_t poly1, const arb_poly_t poly2, slong prec)
void _arb_poly_compose(arb_ptr res, arb_srcptr poly1, slong len1, arb_srcptr poly2, slong len2, slong prec)
void arb_poly_compose(arb_poly_t res, const arb_poly_t poly1, const arb_poly_t poly2, slong prec)

Sets res to the composition \(h(x) = f(g(x))\) where \(f\) is given by poly1 and \(g\) is given by poly2, respectively using Horner’s rule, divide-and-conquer, and an automatic choice between the two algorithms.

The default algorithm also handles special-form input \(g = ax^n + c\) efficiently by performing a Taylor shift followed by a rescaling.

The underscore methods do not support aliasing of the output with either input polynomial.

void _arb_poly_compose_series_horner(arb_ptr res, arb_srcptr poly1, slong len1, arb_srcptr poly2, slong len2, slong n, slong prec)
void arb_poly_compose_series_horner(arb_poly_t res, const arb_poly_t poly1, const arb_poly_t poly2, slong n, slong prec)
void _arb_poly_compose_series_brent_kung(arb_ptr res, arb_srcptr poly1, slong len1, arb_srcptr poly2, slong len2, slong n, slong prec)
void arb_poly_compose_series_brent_kung(arb_poly_t res, const arb_poly_t poly1, const arb_poly_t poly2, slong n, slong prec)
void _arb_poly_compose_series(arb_ptr res, arb_srcptr poly1, slong len1, arb_srcptr poly2, slong len2, slong n, slong prec)
void arb_poly_compose_series(arb_poly_t res, const arb_poly_t poly1, const arb_poly_t poly2, slong n, slong prec)

Sets res to the power series composition \(h(x) = f(g(x))\) truncated to order \(O(x^n)\) where \(f\) is given by poly1 and \(g\) is given by poly2, respectively using Horner’s rule, the Brent-Kung baby step-giant step algorithm, and an automatic choice between the two algorithms.

The default algorithm also handles special-form input \(g = ax^n\) efficiently.

We require that the constant term in \(g(x)\) is exactly zero. The underscore methods do not support aliasing of the output with either input polynomial.

void _arb_poly_revert_series_lagrange(arb_ptr h, arb_srcptr f, slong flen, slong n, slong prec)
void arb_poly_revert_series_lagrange(arb_poly_t h, const arb_poly_t f, slong n, slong prec)
void _arb_poly_revert_series_newton(arb_ptr h, arb_srcptr f, slong flen, slong n, slong prec)
void arb_poly_revert_series_newton(arb_poly_t h, const arb_poly_t f, slong n, slong prec)
void _arb_poly_revert_series_lagrange_fast(arb_ptr h, arb_srcptr f, slong flen, slong n, slong prec)
void arb_poly_revert_series_lagrange_fast(arb_poly_t h, const arb_poly_t f, slong n, slong prec)
void _arb_poly_revert_series(arb_ptr h, arb_srcptr f, slong flen, slong n, slong prec)
void arb_poly_revert_series(arb_poly_t h, const arb_poly_t f, slong n, slong prec)

Sets \(h\) to the power series reversion of \(f\), i.e. the expansion of the compositional inverse function \(f^{-1}(x)\), truncated to order \(O(x^n)\), using respectively Lagrange inversion, Newton iteration, fast Lagrange inversion, and a default algorithm choice.

We require that the constant term in \(f\) is exactly zero and that the linear term is nonzero. The underscore methods assume that flen is at least 2, and do not support aliasing.

Evaluation

void _arb_poly_evaluate_horner(arb_t y, arb_srcptr f, slong len, const arb_t x, slong prec)
void arb_poly_evaluate_horner(arb_t y, const arb_poly_t f, const arb_t x, slong prec)
void _arb_poly_evaluate_rectangular(arb_t y, arb_srcptr f, slong len, const arb_t x, slong prec)
void arb_poly_evaluate_rectangular(arb_t y, const arb_poly_t f, const arb_t x, slong prec)
void _arb_poly_evaluate(arb_t y, arb_srcptr f, slong len, const arb_t x, slong prec)
void arb_poly_evaluate(arb_t y, const arb_poly_t f, const arb_t x, slong prec)

Sets \(y = f(x)\), evaluated respectively using Horner’s rule, rectangular splitting, and an automatic algorithm choice.

void _arb_poly_evaluate_acb_horner(acb_t y, arb_srcptr f, slong len, const acb_t x, slong prec)
void arb_poly_evaluate_acb_horner(acb_t y, const arb_poly_t f, const acb_t x, slong prec)
void _arb_poly_evaluate_acb_rectangular(acb_t y, arb_srcptr f, slong len, const acb_t x, slong prec)
void arb_poly_evaluate_acb_rectangular(acb_t y, const arb_poly_t f, const acb_t x, slong prec)
void _arb_poly_evaluate_acb(acb_t y, arb_srcptr f, slong len, const acb_t x, slong prec)
void arb_poly_evaluate_acb(acb_t y, const arb_poly_t f, const acb_t x, slong prec)

Sets \(y = f(x)\) where \(x\) is a complex number, evaluating the polynomial respectively using Horner’s rule, rectangular splitting, and an automatic algorithm choice.

void _arb_poly_evaluate2_horner(arb_t y, arb_t z, arb_srcptr f, slong len, const arb_t x, slong prec)
void arb_poly_evaluate2_horner(arb_t y, arb_t z, const arb_poly_t f, const arb_t x, slong prec)
void _arb_poly_evaluate2_rectangular(arb_t y, arb_t z, arb_srcptr f, slong len, const arb_t x, slong prec)
void arb_poly_evaluate2_rectangular(arb_t y, arb_t z, const arb_poly_t f, const arb_t x, slong prec)
void _arb_poly_evaluate2(arb_t y, arb_t z, arb_srcptr f, slong len, const arb_t x, slong prec)
void arb_poly_evaluate2(arb_t y, arb_t z, const arb_poly_t f, const arb_t x, slong prec)

Sets \(y = f(x), z = f'(x)\), evaluated respectively using Horner’s rule, rectangular splitting, and an automatic algorithm choice.

When Horner’s rule is used, the only advantage of evaluating the function and its derivative simultaneously is that one does not have to generate the derivative polynomial explicitly. With the rectangular splitting algorithm, the powers can be reused, making simultaneous evaluation slightly faster.

void _arb_poly_evaluate2_acb_horner(acb_t y, acb_t z, arb_srcptr f, slong len, const acb_t x, slong prec)
void arb_poly_evaluate2_acb_horner(acb_t y, acb_t z, const arb_poly_t f, const acb_t x, slong prec)
void _arb_poly_evaluate2_acb_rectangular(acb_t y, acb_t z, arb_srcptr f, slong len, const acb_t x, slong prec)
void arb_poly_evaluate2_acb_rectangular(acb_t y, acb_t z, const arb_poly_t f, const acb_t x, slong prec)
void _arb_poly_evaluate2_acb(acb_t y, acb_t z, arb_srcptr f, slong len, const acb_t x, slong prec)
void arb_poly_evaluate2_acb(acb_t y, acb_t z, const arb_poly_t f, const acb_t x, slong prec)

Sets \(y = f(x), z = f'(x)\), evaluated respectively using Horner’s rule, rectangular splitting, and an automatic algorithm choice.

Product trees

void _arb_poly_product_roots(arb_ptr poly, arb_srcptr xs, slong n, slong prec)
void arb_poly_product_roots(arb_poly_t poly, arb_srcptr xs, slong n, slong prec)

Generates the polynomial \((x-x_0)(x-x_1)\cdots(x-x_{n-1})\).

void _arb_poly_product_roots_complex(arb_ptr poly, arb_srcptr r, slong rn, acb_srcptr c, slong cn, slong prec)
void arb_poly_product_roots_complex(arb_poly_t poly, arb_srcptr r, slong rn, acb_srcptr c, slong cn, slong prec)

Generates the polynomial

\[\left(\prod_{i=0}^{rn-1} (x-r_i)\right) \left(\prod_{i=0}^{cn-1} (x-c_i)(x-\bar{c_i})\right)\]

having rn real roots given by the array r and having \(2cn\) complex roots in conjugate pairs given by the length-cn array c. Either rn or cn or both may be zero.

Note that only one representative from each complex conjugate pair is supplied (unless a pair is supposed to be repeated with higher multiplicity). To construct a polynomial from complex roots where the conjugate pairs have not been distinguished, use acb_poly_product_roots() instead.

arb_ptr * _arb_poly_tree_alloc(slong len)

Returns an initialized data structured capable of representing a remainder tree (product tree) of len roots.

void _arb_poly_tree_free(arb_ptr * tree, slong len)

Deallocates a tree structure as allocated using _arb_poly_tree_alloc.

void _arb_poly_tree_build(arb_ptr * tree, arb_srcptr roots, slong len, slong prec)

Constructs a product tree from a given array of len roots. The tree structure must be pre-allocated to the specified length using _arb_poly_tree_alloc().

Multipoint evaluation

void _arb_poly_evaluate_vec_iter(arb_ptr ys, arb_srcptr poly, slong plen, arb_srcptr xs, slong n, slong prec)
void arb_poly_evaluate_vec_iter(arb_ptr ys, const arb_poly_t poly, arb_srcptr xs, slong n, slong prec)

Evaluates the polynomial simultaneously at n given points, calling _arb_poly_evaluate() repeatedly.

void _arb_poly_evaluate_vec_fast_precomp(arb_ptr vs, arb_srcptr poly, slong plen, arb_ptr * tree, slong len, slong prec)
void _arb_poly_evaluate_vec_fast(arb_ptr ys, arb_srcptr poly, slong plen, arb_srcptr xs, slong n, slong prec)
void arb_poly_evaluate_vec_fast(arb_ptr ys, const arb_poly_t poly, arb_srcptr xs, slong n, slong prec)

Evaluates the polynomial simultaneously at n given points, using fast multipoint evaluation.

Interpolation

void _arb_poly_interpolate_newton(arb_ptr poly, arb_srcptr xs, arb_srcptr ys, slong n, slong prec)
void arb_poly_interpolate_newton(arb_poly_t poly, arb_srcptr xs, arb_srcptr ys, slong n, slong prec)

Recovers the unique polynomial of length at most n that interpolates the given x and y values. This implementation first interpolates in the Newton basis and then converts back to the monomial basis.

void _arb_poly_interpolate_barycentric(arb_ptr poly, arb_srcptr xs, arb_srcptr ys, slong n, slong prec)
void arb_poly_interpolate_barycentric(arb_poly_t poly, arb_srcptr xs, arb_srcptr ys, slong n, slong prec)

Recovers the unique polynomial of length at most n that interpolates the given x and y values. This implementation uses the barycentric form of Lagrange interpolation.

void _arb_poly_interpolation_weights(arb_ptr w, arb_ptr * tree, slong len, slong prec)
void _arb_poly_interpolate_fast_precomp(arb_ptr poly, arb_srcptr ys, arb_ptr * tree, arb_srcptr weights, slong len, slong prec)
void _arb_poly_interpolate_fast(arb_ptr poly, arb_srcptr xs, arb_srcptr ys, slong len, slong prec)
void arb_poly_interpolate_fast(arb_poly_t poly, arb_srcptr xs, arb_srcptr ys, slong n, slong prec)

Recovers the unique polynomial of length at most n that interpolates the given x and y values, using fast Lagrange interpolation. The precomp function takes a precomputed product tree over the x values and a vector of interpolation weights as additional inputs.

Differentiation

void _arb_poly_derivative(arb_ptr res, arb_srcptr poly, slong len, slong prec)

Sets {res, len - 1} to the derivative of {poly, len}. Allows aliasing of the input and output.

void arb_poly_derivative(arb_poly_t res, const arb_poly_t poly, slong prec)

Sets res to the derivative of poly.

void _arb_poly_integral(arb_ptr res, arb_srcptr poly, slong len, slong prec)

Sets {res, len} to the integral of {poly, len - 1}. Allows aliasing of the input and output.

void arb_poly_integral(arb_poly_t res, const arb_poly_t poly, slong prec)

Sets res to the integral of poly.

Transforms

void _arb_poly_borel_transform(arb_ptr res, arb_srcptr poly, slong len, slong prec)
void arb_poly_borel_transform(arb_poly_t res, const arb_poly_t poly, slong prec)

Computes the Borel transform of the input polynomial, mapping \(\sum_k a_k x^k\) to \(\sum_k (a_k / k!) x^k\). The underscore method allows aliasing.

void _arb_poly_inv_borel_transform(arb_ptr res, arb_srcptr poly, slong len, slong prec)
void arb_poly_inv_borel_transform(arb_poly_t res, const arb_poly_t poly, slong prec)

Computes the inverse Borel transform of the input polynomial, mapping \(\sum_k a_k x^k\) to \(\sum_k a_k k! x^k\). The underscore method allows aliasing.

void _arb_poly_binomial_transform_basecase(arb_ptr b, arb_srcptr a, slong alen, slong len, slong prec)
void arb_poly_binomial_transform_basecase(arb_poly_t b, const arb_poly_t a, slong len, slong prec)
void _arb_poly_binomial_transform_convolution(arb_ptr b, arb_srcptr a, slong alen, slong len, slong prec)
void arb_poly_binomial_transform_convolution(arb_poly_t b, const arb_poly_t a, slong len, slong prec)
void _arb_poly_binomial_transform(arb_ptr b, arb_srcptr a, slong alen, slong len, slong prec)
void arb_poly_binomial_transform(arb_poly_t b, const arb_poly_t a, slong len, slong prec)

Computes the binomial transform of the input polynomial, truncating the output to length len. The binomial transform maps the coefficients \(a_k\) in the input polynomial to the coefficients \(b_k\) in the output polynomial via \(b_n = \sum_{k=0}^n (-1)^k {n \choose k} a_k\). The binomial transform is equivalent to the power series composition \(f(x) \to (1-x)^{-1} f(x/(x-1))\), and is its own inverse.

The basecase version evaluates coefficients one by one from the definition, generating the binomial coefficients by a recurrence relation.

The convolution version uses the identity \(T(f(x)) = B^{-1}(e^x B(f(-x)))\) where \(T\) denotes the binomial transform operator and \(B\) denotes the Borel transform operator. This only costs a single polynomial multiplication, plus some scalar operations.

The default version automatically chooses an algorithm.

The underscore methods do not support aliasing, and assume that the lengths are nonzero.

Powers and elementary functions

void _arb_poly_pow_ui_trunc_binexp(arb_ptr res, arb_srcptr f, slong flen, ulong exp, slong len, slong prec)

Sets {res, len} to {f, flen} raised to the power exp, truncated to length len. Requires that len is no longer than the length of the power as computed without truncation (i.e. no zero-padding is performed). Does not support aliasing of the input and output, and requires that flen and len are positive. Uses binary expontiation.

void arb_poly_pow_ui_trunc_binexp(arb_poly_t res, const arb_poly_t poly, ulong exp, slong len, slong prec)

Sets res to poly raised to the power exp, truncated to length len. Uses binary exponentiation.

void _arb_poly_pow_ui(arb_ptr res, arb_srcptr f, slong flen, ulong exp, slong prec)

Sets res to {f, flen} raised to the power exp. Does not support aliasing of the input and output, and requires that flen is positive.

void arb_poly_pow_ui(arb_poly_t res, const arb_poly_t poly, ulong exp, slong prec)

Sets res to poly raised to the power exp.

void _arb_poly_pow_series(arb_ptr h, arb_srcptr f, slong flen, arb_srcptr g, slong glen, slong len, slong prec)

Sets {h, len} to the power series \(f(x)^{g(x)} = \exp(g(x) \log f(x))\) truncated to length len. This function detects special cases such as g being an exact small integer or \(\pm 1/2\), and computes such powers more efficiently. This function does not support aliasing of the output with either of the input operands. It requires that all lengths are positive, and assumes that flen and glen do not exceed len.

void arb_poly_pow_series(arb_poly_t h, const arb_poly_t f, const arb_poly_t g, slong len, slong prec)

Sets h to the power series \(f(x)^{g(x)} = \exp(g(x) \log f(x))\) truncated to length len. This function detects special cases such as g being an exact small integer or \(\pm 1/2\), and computes such powers more efficiently.

void _arb_poly_pow_arb_series(arb_ptr h, arb_srcptr f, slong flen, const arb_t g, slong len, slong prec)

Sets {h, len} to the power series \(f(x)^g = \exp(g \log f(x))\) truncated to length len. This function detects special cases such as g being an exact small integer or \(\pm 1/2\), and computes such powers more efficiently. This function does not support aliasing of the output with either of the input operands. It requires that all lengths are positive, and assumes that flen does not exceed len.

void arb_poly_pow_arb_series(arb_poly_t h, const arb_poly_t f, const arb_t g, slong len, slong prec)

Sets h to the power series \(f(x)^g = \exp(g \log f(x))\) truncated to length len.

void _arb_poly_sqrt_series(arb_ptr g, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_sqrt_series(arb_poly_t g, const arb_poly_t h, slong n, slong prec)

Sets g to the power series square root of h, truncated to length n. Uses division-free Newton iteration for the reciprocal square root, followed by a multiplication.

The underscore method does not support aliasing of the input and output arrays. It requires that hlen and n are greater than zero.

void _arb_poly_rsqrt_series(arb_ptr g, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_rsqrt_series(arb_poly_t g, const arb_poly_t h, slong n, slong prec)

Sets g to the reciprocal power series square root of h, truncated to length n. Uses division-free Newton iteration.

The underscore method does not support aliasing of the input and output arrays. It requires that hlen and n are greater than zero.

void _arb_poly_log_series(arb_ptr res, arb_srcptr f, slong flen, slong n, slong prec)
void arb_poly_log_series(arb_poly_t res, const arb_poly_t f, slong n, slong prec)

Sets res to the power series logarithm of f, truncated to length n. Uses the formula \(\log(f(x)) = \int f'(x) / f(x) dx\), adding the logarithm of the constant term in f as the constant of integration.

The underscore method supports aliasing of the input and output arrays. It requires that flen and n are greater than zero.

void _arb_poly_log1p_series(arb_ptr res, arb_srcptr f, slong flen, slong n, slong prec)
void arb_poly_log1p_series(arb_poly_t res, const arb_poly_t f, slong n, slong prec)

Computes the power series \(\log(1+f)\), with better accuracy when the constant term of f is small.

void _arb_poly_atan_series(arb_ptr res, arb_srcptr f, slong flen, slong n, slong prec)
void arb_poly_atan_series(arb_poly_t res, const arb_poly_t f, slong n, slong prec)
void _arb_poly_asin_series(arb_ptr res, arb_srcptr f, slong flen, slong n, slong prec)
void arb_poly_asin_series(arb_poly_t res, const arb_poly_t f, slong n, slong prec)
void _arb_poly_acos_series(arb_ptr res, arb_srcptr f, slong flen, slong n, slong prec)
void arb_poly_acos_series(arb_poly_t res, const arb_poly_t f, slong n, slong prec)

Sets res respectively to the power series inverse tangent, inverse sine and inverse cosine of f, truncated to length n.

Uses the formulas

\[ \begin{align}\begin{aligned}\tan^{-1}(f(x)) = \int f'(x) / (1+f(x)^2) dx,\\\sin^{-1}(f(x)) = \int f'(x) / (1-f(x)^2)^{1/2} dx,\\\cos^{-1}(f(x)) = -\int f'(x) / (1-f(x)^2)^{1/2} dx,\end{aligned}\end{align} \]

adding the inverse function of the constant term in f as the constant of integration.

The underscore methods supports aliasing of the input and output arrays. They require that flen and n are greater than zero.

void _arb_poly_exp_series_basecase(arb_ptr f, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_exp_series_basecase(arb_poly_t f, const arb_poly_t h, slong n, slong prec)
void _arb_poly_exp_series(arb_ptr f, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_exp_series(arb_poly_t f, const arb_poly_t h, slong n, slong prec)

Sets \(f\) to the power series exponential of \(h\), truncated to length \(n\).

The basecase version uses a simple recurrence for the coefficients, requiring \(O(nm)\) operations where \(m\) is the length of \(h\).

The main implementation uses Newton iteration, starting from a small number of terms given by the basecase algorithm. The complexity is \(O(M(n))\). Redundant operations in the Newton iteration are avoided by using the scheme described in [HZ2004].

The underscore methods support aliasing and allow the input to be shorter than the output, but require the lengths to be nonzero.

void _arb_poly_sin_cos_series_basecase(arb_ptr s, arb_ptr c, arb_srcptr h, slong hlen, slong n, slong prec, int times_pi)
void arb_poly_sin_cos_series_basecase(arb_poly_t s, arb_poly_t c, const arb_poly_t h, slong n, slong prec, int times_pi)
void _arb_poly_sin_cos_series_tangent(arb_ptr s, arb_ptr c, arb_srcptr h, slong hlen, slong n, slong prec, int times_pi)
void arb_poly_sin_cos_series_tangent(arb_poly_t s, arb_poly_t c, const arb_poly_t h, slong n, slong prec, int times_pi)
void _arb_poly_sin_cos_series(arb_ptr s, arb_ptr c, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_sin_cos_series(arb_poly_t s, arb_poly_t c, const arb_poly_t h, slong n, slong prec)

Sets s and c to the power series sine and cosine of h, computed simultaneously.

The basecase version uses a simple recurrence for the coefficients, requiring \(O(nm)\) operations where \(m\) is the length of \(h\).

The tangent version uses the tangent half-angle formulas to compute the sine and cosine via _arb_poly_tan_series(). This requires \(O(M(n))\) operations. When \(h = h_0 + h_1\) where the constant term \(h_0\) is nonzero, the evaluation is done as \(\sin(h_0 + h_1) = \cos(h_0) \sin(h_1) + \sin(h_0) \cos(h_1)\), \(\cos(h_0 + h_1) = \cos(h_0) \cos(h_1) - \sin(h_0) \sin(h_1)\), to improve accuracy and avoid dividing by zero at the poles of the tangent function.

The default version automatically selects between the basecase and tangent algorithms depending on the input.

The basecase and tangent versions take a flag times_pi specifying that the input is to be multiplied by \(\pi\).

The underscore methods support aliasing and require the lengths to be nonzero.

void _arb_poly_sin_series(arb_ptr s, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_sin_series(arb_poly_t s, const arb_poly_t h, slong n, slong prec)
void _arb_poly_cos_series(arb_ptr c, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_cos_series(arb_poly_t c, const arb_poly_t h, slong n, slong prec)

Respectively evaluates the power series sine or cosine. These functions simply wrap _arb_poly_sin_cos_series(). The underscore methods support aliasing and require the lengths to be nonzero.

void _arb_poly_tan_series(arb_ptr g, arb_srcptr h, slong hlen, slong len, slong prec)
void arb_poly_tan_series(arb_poly_t g, const arb_poly_t h, slong n, slong prec)

Sets g to the power series tangent of h.

For small n takes the quotient of the sine and cosine as computed using the basecase algorithm. For large n, uses Newton iteration to invert the inverse tangent series. The complexity is \(O(M(n))\).

The underscore version does not support aliasing, and requires the lengths to be nonzero.

void _arb_poly_sin_cos_pi_series(arb_ptr s, arb_ptr c, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_sin_cos_pi_series(arb_poly_t s, arb_poly_t c, const arb_poly_t h, slong n, slong prec)
void _arb_poly_sin_pi_series(arb_ptr s, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_sin_pi_series(arb_poly_t s, const arb_poly_t h, slong n, slong prec)
void _arb_poly_cos_pi_series(arb_ptr c, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_cos_pi_series(arb_poly_t c, const arb_poly_t h, slong n, slong prec)
void _arb_poly_cot_pi_series(arb_ptr c, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_cot_pi_series(arb_poly_t c, const arb_poly_t h, slong n, slong prec)

Compute the respective trigonometric functions of the input multiplied by \(\pi\).

void _arb_poly_sinh_cosh_series_basecase(arb_ptr s, arb_ptr c, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_sinh_cosh_series_basecase(arb_poly_t s, arb_poly_t c, const arb_poly_t h, slong n, slong prec)
void _arb_poly_sinh_cosh_series_exponential(arb_ptr s, arb_ptr c, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_sinh_cosh_series_exponential(arb_poly_t s, arb_poly_t c, const arb_poly_t h, slong n, slong prec)
void _arb_poly_sinh_cosh_series(arb_ptr s, arb_ptr c, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_sinh_cosh_series(arb_poly_t s, arb_poly_t c, const arb_poly_t h, slong n, slong prec)
void _arb_poly_sinh_series(arb_ptr s, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_sinh_series(arb_poly_t s, const arb_poly_t h, slong n, slong prec)
void _arb_poly_cosh_series(arb_ptr c, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_cosh_series(arb_poly_t c, const arb_poly_t h, slong n, slong prec)

Sets s and c respectively to the hyperbolic sine and cosine of the power series h, truncated to length n.

The implementations mirror those for sine and cosine, except that the exponential version computes both functions using the exponential function instead of the hyperbolic tangent.

void _arb_poly_sinc_series(arb_ptr s, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_sinc_series(arb_poly_t s, const arb_poly_t h, slong n, slong prec)

Sets c to the sinc function of the power series h, truncated to length n.

Lambert W function

void _arb_poly_lambertw_series(arb_ptr res, arb_srcptr z, slong zlen, int flags, slong len, slong prec)
void arb_poly_lambertw_series(arb_poly_t res, const arb_poly_t z, int flags, slong len, slong prec)

Sets res to the Lambert W function of the power series z. If flags is 0, the principal branch is computed; if flags is 1, the second real branch \(W_{-1}(z)\) is computed. The underscore method allows aliasing, but assumes that the lengths are nonzero.

Gamma function and factorials

void _arb_poly_gamma_series(arb_ptr res, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_gamma_series(arb_poly_t res, const arb_poly_t h, slong n, slong prec)
void _arb_poly_rgamma_series(arb_ptr res, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_rgamma_series(arb_poly_t res, const arb_poly_t h, slong n, slong prec)
void _arb_poly_lgamma_series(arb_ptr res, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_lgamma_series(arb_poly_t res, const arb_poly_t h, slong n, slong prec)
void _arb_poly_digamma_series(arb_ptr res, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_digamma_series(arb_poly_t res, const arb_poly_t h, slong n, slong prec)

Sets res to the series expansion of \(\Gamma(h(x))\), \(1/\Gamma(h(x))\), or \(\log \Gamma(h(x))\), \(\psi(h(x))\), truncated to length n.

These functions first generate the Taylor series at the constant term of h, and then call _arb_poly_compose_series(). The Taylor coefficients are generated using the Riemann zeta function if the constant term of h is a small integer, and with Stirling’s series otherwise.

The underscore methods support aliasing of the input and output arrays, and require that hlen and n are greater than zero.

void _arb_poly_rising_ui_series(arb_ptr res, arb_srcptr f, slong flen, ulong r, slong trunc, slong prec)
void arb_poly_rising_ui_series(arb_poly_t res, const arb_poly_t f, ulong r, slong trunc, slong prec)

Sets res to the rising factorial \((f) (f+1) (f+2) \cdots (f+r-1)\), truncated to length trunc. The underscore method assumes that flen, r and trunc are at least 1, and does not support aliasing. Uses binary splitting.

Zeta function

void arb_poly_zeta_series(arb_poly_t res, const arb_poly_t s, const arb_t a, int deflate, slong n, slong prec)

Sets res to the Hurwitz zeta function \(\zeta(s,a)\) where \(s\) a power series and \(a\) is a constant, truncated to length n. To evaluate the usual Riemann zeta function, set \(a = 1\).

If deflate is nonzero, evaluates \(\zeta(s,a) + 1/(1-s)\), which is well-defined as a limit when the constant term of \(s\) is 1. In particular, expanding \(\zeta(s,a) + 1/(1-s)\) with \(s = 1+x\) gives the Stieltjes constants

\[\sum_{k=0}^{n-1} \frac{(-1)^k}{k!} \gamma_k(a) x^k.\]

If \(a = 1\), this implementation uses the reflection formula if the midpoint of the constant term of \(s\) is negative.

void _arb_poly_riemann_siegel_theta_series(arb_ptr res, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_riemann_siegel_theta_series(arb_poly_t res, const arb_poly_t h, slong n, slong prec)

Sets res to the series expansion of the Riemann-Siegel theta function

\[\theta(h) = \arg \left(\Gamma\left(\frac{2ih+1}{4}\right)\right) - \frac{\log \pi}{2} h\]

where the argument of the gamma function is chosen continuously as the imaginary part of the log gamma function.

The underscore method does not support aliasing of the input and output arrays, and requires that the lengths are greater than zero.

void _arb_poly_riemann_siegel_z_series(arb_ptr res, arb_srcptr h, slong hlen, slong n, slong prec)
void arb_poly_riemann_siegel_z_series(arb_poly_t res, const arb_poly_t h, slong n, slong prec)

Sets res to the series expansion of the Riemann-Siegel Z-function

\[Z(h) = e^{i\theta(h)} \zeta(1/2+ih).\]

The zeros of the Z-function on the real line precisely correspond to the imaginary parts of the zeros of the Riemann zeta function on the critical line.

The underscore method supports aliasing of the input and output arrays, and requires that the lengths are greater than zero.

Root-finding

void _arb_poly_root_bound_fujiwara(mag_t bound, arb_srcptr poly, slong len)
void arb_poly_root_bound_fujiwara(mag_t bound, arb_poly_t poly)

Sets bound to an upper bound for the magnitude of all the complex roots of poly. Uses Fujiwara’s bound

\[2 \max \left\{\left|\frac{a_{n-1}}{a_n}\right|, \left|\frac{a_{n-2}}{a_n}\right|^{1/2}, \cdots, \left|\frac{a_1}{a_n}\right|^{1/(n-1)}, \left|\frac{a_0}{2a_n}\right|^{1/n} \right\}\]

where \(a_0, \ldots, a_n\) are the coefficients of poly.

void _arb_poly_newton_convergence_factor(arf_t convergence_factor, arb_srcptr poly, slong len, const arb_t convergence_interval, slong prec)

Given an interval \(I\) specified by convergence_interval, evaluates a bound for \(C = \sup_{t,u \in I} \frac{1}{2} |f''(t)| / |f'(u)|\), where \(f\) is the polynomial defined by the coefficients {poly, len}. The bound is obtained by evaluating \(f'(I)\) and \(f''(I)\) directly. If \(f\) has large coefficients, \(I\) must be extremely precise in order to get a finite factor.

int _arb_poly_newton_step(arb_t xnew, arb_srcptr poly, slong len, const arb_t x, const arb_t convergence_interval, const arf_t convergence_factor, slong prec)

Performs a single step with Newton’s method.

The input consists of the polynomial \(f\) specified by the coefficients {poly, len}, an interval \(x = [m-r, m+r]\) known to contain a single root of \(f\), an interval \(I\) (convergence_interval) containing \(x\) with an associated bound (convergence_factor) for \(C = \sup_{t,u \in I} \frac{1}{2} |f''(t)| / |f'(u)|\), and a working precision prec.

The Newton update consists of setting \(x' = [m'-r', m'+r']\) where \(m' = m - f(m) / f'(m)\) and \(r' = C r^2\). The expression \(m - f(m) / f'(m)\) is evaluated using ball arithmetic at a working precision of prec bits, and the rounding error during this evaluation is accounted for in the output. We now check that \(x' \in I\) and \(m' < m\). If both conditions are satisfied, we set xnew to \(x'\) and return nonzero. If either condition fails, we set xnew to \(x\) and return zero, indicating that no progress was made.

void _arb_poly_newton_refine_root(arb_t r, arb_srcptr poly, slong len, const arb_t start, const arb_t convergence_interval, const arf_t convergence_factor, slong eval_extra_prec, slong prec)

Refines a precise estimate of a polynomial root to high precision by performing several Newton steps, using nearly optimally chosen doubling precision steps.

The inputs are defined as for _arb_poly_newton_step, except for the precision parameters: prec is the target accuracy and eval_extra_prec is the estimated number of guard bits that need to be added to evaluate the polynomial accurately close to the root (typically, if the polynomial has large coefficients of alternating signs, this needs to be approximately the bit size of the coefficients).

Other special polynomials

void _arb_poly_swinnerton_dyer_ui(arb_ptr poly, ulong n, slong trunc, slong prec)
void arb_poly_swinnerton_dyer_ui(arb_poly_t poly, ulong n, slong prec)

Computes the Swinnerton-Dyer polynomial \(S_n\), which has degree \(2^n\) and is the rational minimal polynomial of the sum of the square roots of the first n prime numbers.

If prec is set to zero, a precision is chosen automatically such that arb_poly_get_unique_fmpz_poly() should be successful. Otherwise a working precision of prec bits is used.

The underscore version accepts an additional trunc parameter. Even when computing a truncated polynomial, the array poly must have room for \(2^n + 1\) coefficients, used as temporary space.