LongbitCount

LongbitCount refers to the operation of counting the number of set bits in a 64-bit integer (a long). It is also known as the population count or Hamming weight, representing how many 1s appear in the binary form of the value. In signed representations, the count includes all 64 bits, since they are stored in two’s complement form.

In many programming languages this operation is available as a built-in function. For example, Java provides Long.bitCount(long i), which returns the number of one-bits in i. Other languages offer similar facilities, such as Go’s bits.OnesCount64 or C/C++’s __builtin_popcountll, with variations across language syntax and integer width.

Implementation approaches vary. A straightforward method tests each bit and accumulates the count, resulting in O(n) time for n bits (typically 64). More efficient algorithms exist, such as Kernighan’s method, which repeatedly clears the least-significant set bit, performing as many iterations as there are set bits. The SWAR (SIMD Within A Register) technique uses parallel bit counting to reduce operations further. Modern compilers and processors often map this operation to a dedicated hardware popcount instruction, if available, delivering very fast performance.

Edge cases include negative values, where the sign bit contributes to the total count, and zero, which yields zero. Applications of longbitCount include cryptography, data compression, bitmap processing, and algorithms that require set cardinality or feature counting.

See also: population count, Hamming weight.