Module core::core_arch::wasm32::relaxed_simd
source · 🔬This is a nightly-only experimental API. (
stdsimd #48556)Available on
target_family="wasm" only.Functions
- Computes
a * b + cwith either one rounding or two roundings. - A relaxed version of
f32x4_maxwhich is eitherf32x4_maxorf32x4_pmax. - A relaxed version of
f32x4_minwhich is eitherf32x4_minorf32x4_pmin. - Computes
-a * b + cwith either one rounding or two roundings. - Computes
a * b + cwith either one rounding or two roundings. - A relaxed version of
f64x2_maxwhich is eitherf64x2_maxorf64x2_pmax. - A relaxed version of
f64x2_minwhich is eitherf64x2_minorf64x2_pmin. - Computes
-a * b + cwith either one rounding or two roundings. - A relaxed version of
v128_bitselectwhere this either behaves the same asv128_bitselector the high bit of each lanemis inspected and the corresponding lane ofais chosen if the bit is 1 or the lane ofbis chosen if it’s zero. - A relaxed version of
i8x16_swizzle(a, s)which selects lanes fromausing indices ins. - A relaxed dot-product instruction.
- A relaxed version of
v128_bitselectwhere this either behaves the same asv128_bitselector the high bit of each lanemis inspected and the corresponding lane ofais chosen if the bit is 1 or the lane ofbis chosen if it’s zero. - A relaxed version of
i16x8_relaxed_q15mulrwhere if both lanes arei16::MINthen the result is eitheri16::MINori16::MAX. - Similar to
i16x8_relaxed_dot_i8x16_i7x16except that the intermediatei16x8result is fed intoi32x4_extadd_pairwise_i16x8followed byi32x4_addto add the valuecto the result. - A relaxed version of
v128_bitselectwhere this either behaves the same asv128_bitselector the high bit of each lanemis inspected and the corresponding lane ofais chosen if the bit is 1 or the lane ofbis chosen if it’s zero. - A relaxed version of
i32x4_trunc_sat_f32x4(a)converts thef32lanes ofato signed 32-bit integers. - A relaxed version of
i32x4_trunc_sat_f64x2_zero(a)converts thef64lanes ofato signed 32-bit integers and the upper two lanes are zero. - A relaxed version of
v128_bitselectwhere this either behaves the same asv128_bitselector the high bit of each lanemis inspected and the corresponding lane ofais chosen if the bit is 1 or the lane ofbis chosen if it’s zero. - A relaxed version of
u32x4_trunc_sat_f32x4(a)converts thef32lanes ofato unsigned 32-bit integers. - A relaxed version of
u32x4_trunc_sat_f64x2_zero(a)converts thef64lanes ofato unsigned 32-bit integers and the upper two lanes are zero.