Nekotekina 99b5284463 X86: avoid vector-scalar shifts if splat amount is directly a vector ADD/SUB/AND op.
Prefer vector-vector shifts if available (AVX2+).
Improves code generated for rotate and funnel shifts.
Otherwise it would generate a shuffle + slower vector-scalar shift.
2019-04-21 22:35:31 +03:00
..
2019-04-12 02:02:06 +00:00
2019-04-12 02:02:06 +00:00
2019-04-12 02:02:06 +00:00
2019-04-05 08:26:58 +00:00