log2_ceil!
Static base 2 logarithm of a scalar, rounded up. Integer and field arithmetic
log2_ceil! n
Returns ceil(log2(n)) for a static scalar n taken as its canonical integer: the number of bits needed to index n items, or the number of halvings that reduce n to one.
Arguments
n- Static scalar.
Returns
A static scalar. Expression form only.
Constraints
None.
Example
(let x) static a = log2_ceil! 5 static b = log2_ceil! 1 static c = log2_ceil! 4 x * [a b c]
Notes
Computed in f64, which is exact for the sizes used in practice but can misround for values very close to a power of two above 2^53. n = 0 gives 0. The argument runs to the next comma: log2_ceil! k + 1 is the logarithm of k + 1, so wrap the precompile in parentheses inside larger expressions (16 + (log2_ceil! terms)). Errors: input must be static, input must be scalar.
Used in the standard library
bits/add32_bits, blake3/blake3_core, mersenne61/mldsa/mldsa_reduce, milli/mldsa/mldsa_reduce, polynomial/poly_norm_inf_lt, slip/slip_r1cs_arg_verify