MCPcopy Create free account
hub / github.com/cwida/ALP / sse_int64_to_double

Function sse_int64_to_double

include/alp/decoder.hpp:59–62  ·  view source on GitHub ↗

SSE version of int64_to_double Only works for inputs in the range: [-2^51, 2^51]

Source from the content-addressed store, hash-verified

57// SSE version of int64_to_double
58// Only works for inputs in the range: [-2^51, 2^51]
59__m128d sse_int64_to_double(__m128i x) {
60 x = _mm_add_epi64(x, _mm_castpd_si128(_mm_set1_pd(0x0018000000000000)));
61 return _mm_sub_pd(_mm_castsi128_pd(x), _mm_set1_pd(0x0018000000000000));
62}
63
64__m256d int64_to_double_fast_precise(const __m256i v)
65/* Optimized full range int64_t to double conversion */

Callers 1

sse_decodeFunction · 0.85

Calls

no outgoing calls

Tested by

no test coverage detected