pub unsafe fn _mm256_slli_si256(a: __m256i, const IMM8: i32) -> __m256i
avx2
Shifts 128-bit lanes in a left by imm8 bytes while shifting in zeros.
a
imm8
Intel’s documentation