WebAssembly SIMD conversion instructions.
convert_low_i32x4_sConverts the lanes of a v128 f64x2 value interpretation into a signed i32x4 value interpretation.
convert_low_i32x4_uConverts the lanes of a v128 f64x2 value interpretation into an unsigned i32x4 value interpretation.
convert_i32x4_sConverts the lanes of a v128 f32x4 value interpretation into a signed i32x4 value interpretation.
convert_i32x4_uConverts the lanes of a v128 f32x4 value interpretation into an unsigned i32x4 value interpretation.
demote_f64x2_zeroConverts the lanes of a v128 f64x2 value interpretation into an f32x4 value interpretation. The two higher lanes of the result are initialized to zero.
promote_low_f32x4Converts the first two lanes of a v128 f32x4 value interpretation into an f64x2 value interpretation.
extend_high_i8x16_sConverts lanes 8–15 of a signed v128 i8x16 value interpretation into an i16x8 value interpretation.
extend_high_i8x16_uConverts lanes 8–15 of an unsigned v128 i8x16 value interpretation into an i16x8 value interpretation.
extend_high_i16x8_sConverts lanes 4–7 of a signed v128 i16x8 value interpretation into an i32x4 value interpretation.
extend_high_i16x8_uConverts lanes 4–7 of an unsigned v128 i16x8 value interpretation into an i32x4 value interpretation.
extend_high_i32x4_sConverts lanes 2–3 of a signed v128 i32x4 value interpretation into an i64x2 value interpretation.
extend_high_i32x4_uConverts lanes 2–3 of an unsigned v128 i32x4 value interpretation into an i64x2 value interpretation.
extend_low_i8x16_sConverts lanes 0–7 of a signed v128 i8x16 value interpretation into an i16x8 value interpretation.
extend_low_i8x16_uConverts lanes 0–7 of an unsigned v128 i8x16 value interpretation into an i16x8 value interpretation.
extend_low_i16x8_sConverts lanes 0–3 of a signed v128 i16x8 value interpretation into an i32x4 value interpretation.
extend_low_i16x8_uConverts lanes 0–3 of an unsigned v128 i16x8 value interpretation into an i32x4 value interpretation.
extend_low_i32x4_sConverts lanes 0–1 of a signed v128 i32x4 value interpretation into an i64x2 value interpretation.
extend_low_i32x4_uConverts lanes 0–1 of an unsigned v128 i32x4 value interpretation into an i64x2 value interpretation.
narrow_i16x8_sConverts two signed v128 i16x8 value interpretations into an i8x16 value interpretation.
narrow_i16x8_uConverts two signed v128 i16x8 value interpretations into an i8x16 value interpretation using unsigned saturation.
narrow_i32x4_sConverts two signed v128 i32x4 value interpretations into an i16x8 value interpretation.
narrow_i32x4_uConverts two signed v128 i32x4 value interpretations into an i16x8 value interpretation using unsigned saturation.
replace_laneReplaces the specified lane of a v128 value interpretation with a new value and returns the resulting v128 value interpretation.
shuffleReturns a new v128 value interpretation with its lane values selected from two input v128 values, determined by provided index values.
splatCopies the same value to all lanes of a v128 value interpretation.
swizzleReturns a new v128 value interpretation with its lane values selected from an input v128 value, determined by indices provided in a second input v128.
trunc_sat_f32x4_sPerforms a saturating conversion of the lanes of a v128 f32x4 value interpretation into a signed i32x4 value interpretation, clamping the output to the range allowed by the value type.
trunc_sat_f32x4_uPerforms a saturating conversion of the lanes of a v128 f32x4 value interpretation into an unsigned i32x4 value interpretation, clamping the output to the range allowed by the value type.
trunc_sat_f64x2_s_zeroPerforms a saturating conversion of the lanes of a v128 f64x2 value interpretation into a signed i32x4 value interpretation, clamping the output to the range allowed by the value type. The two higher lanes of the result are initialized to zero.
trunc_sat_f64x2_u_zeroPerforms a saturating conversion of the lanes of a v128 f64x2 value interpretation into an unsigned i32x4 value interpretation, clamping the output to the range allowed by the value type. The two higher lanes of the result are initialized to zero.
Note: See also the non-SIMD-specific trunc instruction.