// Code generated by 'tmplgen'; DO NOT EDIT.

//go:build goexperiment.simd

package archsimd

// TruncToInt8 converts each element of x to int8 by discarding its high
// bits.
//
//	z[i] = int8(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Int16x8) TruncToInt8() (z Int8x16) {
	return x.ToBits().TruncToUint8().BitsToInt8()
}

// TruncToUint8 converts each element of x to uint8 by discarding its high
// bits.
//
//	z[i] = uint8(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Uint16x8) TruncToUint8() (z Uint8x16) {
	return x.ReshapeToUint8s().LookupModOrZero(LoadInt8x16Array(&trunc16To8Idx))
}

// TruncToInt16 converts each element of x to int16 by discarding its high
// bits.
//
//	z[i] = int16(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Int32x4) TruncToInt16() (z Int16x8) {
	return x.ToBits().TruncToUint16().BitsToInt16()
}

// TruncToUint16 converts each element of x to uint16 by discarding its high
// bits.
//
//	z[i] = uint16(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Uint32x4) TruncToUint16() (z Uint16x8) {
	return x.ReshapeToUint8s().LookupModOrZero(LoadInt8x16Array(&trunc32To16Idx)).ReshapeToUint16s()
}

// TruncToInt32 converts each element of x to int32 by discarding its high
// bits.
//
//	z[i] = int32(x[i])
//
// The results are stored in the low 2 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Int64x2) TruncToInt32() (z Int32x4) {
	return x.ToBits().TruncToUint32().BitsToInt32()
}

// TruncToUint32 converts each element of x to uint32 by discarding its high
// bits.
//
//	z[i] = uint32(x[i])
//
// The results are stored in the low 2 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Uint64x2) TruncToUint32() (z Uint32x4) {
	var zero Uint32x4
	return x.ReshapeToUint32s().ConcatPermuteScalars(0, 2, 4, 4, zero)
}

// TruncToInt8 converts each element of x to int8 by discarding its high
// bits.
//
//	z[i] = int8(x[i])
//
// The results are stored in the low 16 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Int16x16) TruncToInt8() (z Int8x32) {
	return x.ToBits().TruncToUint8().BitsToInt8()
}

// TruncToUint8 converts each element of x to uint8 by discarding its high
// bits.
//
//	z[i] = uint8(x[i])
//
// The results are stored in the low 16 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Uint16x16) TruncToUint8() (z Uint8x32) {
	if X86.AVX512() {
		return x.truncToUint8()
	}
	g := x.ReshapeToUint8s().LookupModOrZeroGrouped(LoadInt8x32Array(&trunc16To8GroupedIdx))
	return gatherLo64(g.ReshapeToUint64s()).ReshapeToUint8s()
}

// TruncToInt16 converts each element of x to int16 by discarding its high
// bits.
//
//	z[i] = int16(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Int32x8) TruncToInt16() (z Int16x16) {
	return x.ToBits().TruncToUint16().BitsToInt16()
}

// TruncToUint16 converts each element of x to uint16 by discarding its high
// bits.
//
//	z[i] = uint16(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Uint32x8) TruncToUint16() (z Uint16x16) {
	if X86.AVX512() {
		return x.truncToUint16()
	}
	g := x.ReshapeToUint8s().LookupModOrZeroGrouped(LoadInt8x32Array(&trunc32To16GroupedIdx))
	return gatherLo64(g.ReshapeToUint64s()).ReshapeToUint16s()
}

// TruncToInt32 converts each element of x to int32 by discarding its high
// bits.
//
//	z[i] = int32(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Int64x4) TruncToInt32() (z Int32x8) {
	return x.ToBits().TruncToUint32().BitsToInt32()
}

// TruncToUint32 converts each element of x to uint32 by discarding its high
// bits.
//
//	z[i] = uint32(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Uint64x4) TruncToUint32() (z Uint32x8) {
	if X86.AVX512() {
		return x.truncToUint32()
	}
	var zero Uint32x8
	g := x.ReshapeToUint32s().ConcatPermuteScalarsGrouped(0, 2, 4, 4, zero)
	return gatherLo64(g.ReshapeToUint64s()).ReshapeToUint32s()
}

// SaturateToInt8 converts each element of x to int8, saturating values
// that are outside the range of int8.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Int16x8) SaturateToInt8() (z Int8x16) {
	var zero Int16x8
	return x.concatSaturateToInt8(zero)
}

// SaturateToUint8 converts each element of x to uint8, saturating values
// that are outside the range of uint8. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Int16x8) SaturateToUint8() (z Uint8x16) {
	var zero Int16x8
	return x.concatSaturateToUint8(zero)
}

// SaturateToUint8 converts each element of x to uint8, saturating values
// that are outside the range of uint8.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Uint16x8) SaturateToUint8() (z Uint8x16) {
	if X86.AVX512() {
		return x.saturateToUint8()
	}
	// After the Min, x is non-negative as an int16, so the signed-to-unsigned
	// saturation doesn't change it.
	x = x.Min(LoadUint16x8Array(&maxUint8x8))
	return x.BitsToInt16().SaturateToUint8()
}

// SaturateToInt16 converts each element of x to int16, saturating values
// that are outside the range of int16.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Int32x4) SaturateToInt16() (z Int16x8) {
	var zero Int32x4
	return x.ConcatSaturateToInt16(zero)
}

// SaturateToUint16 converts each element of x to uint16, saturating values
// that are outside the range of uint16. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Int32x4) SaturateToUint16() (z Uint16x8) {
	var zero Int32x4
	return x.ConcatSaturateToUint16(zero)
}

// SaturateToUint16 converts each element of x to uint16, saturating values
// that are outside the range of uint16.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Uint32x4) SaturateToUint16() (z Uint16x8) {
	if X86.AVX512() {
		return x.saturateToUint16()
	}
	// After the Min, x is non-negative as an int32, so the signed-to-unsigned
	// saturation doesn't change it.
	x = x.Min(LoadUint32x4Array(&maxUint16x4))
	return x.BitsToInt32().SaturateToUint16()
}

// SaturateToInt32 converts each element of x to int32, saturating values
// that are outside the range of int32.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 2 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Int64x2) SaturateToInt32() (z Int32x4) {
	if X86.AVX512() {
		return x.saturateToInt32()
	}
	hi, lo := LoadInt64x2Array(&maxInt32x2), LoadInt64x2Array(&minInt32x2)
	x = hi.IfElse(x.Greater(hi), x)
	x = lo.IfElse(x.Less(lo), x)
	return x.TruncToInt32()
}

// SaturateToUint32 converts each element of x to uint32, saturating values
// that are outside the range of uint32. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 2 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Int64x2) SaturateToUint32() (z Uint32x4) {
	var zero Int64x2
	if X86.AVX512() {
		return x.Max(zero).ToBits().saturateToUint32()
	}
	hi := LoadInt64x2Array(&maxUint32x2)
	x = x.Masked(x.Greater(zero))
	x = hi.IfElse(x.Greater(hi), x)
	return x.ToBits().TruncToUint32()
}

// SaturateToUint32 converts each element of x to uint32, saturating values
// that are outside the range of uint32.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 2 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX
func (x Uint64x2) SaturateToUint32() (z Uint32x4) {
	if X86.AVX512() {
		return x.saturateToUint32()
	}
	// x fits in a uint32 if its high 32 bits are zero.
	var zero Uint64x2
	hi := LoadInt64x2Array(&maxUint32x2).ToBits()
	return x.IfElse(x.ShiftAllRight(32).Equal(zero), hi).TruncToUint32()
}

// SaturateToInt8 converts each element of x to int8, saturating values
// that are outside the range of int8.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 16 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Int16x16) SaturateToInt8() (z Int8x32) {
	if X86.AVX512() {
		return x.saturateToInt8()
	}
	var zero Int16x16
	g := x.concatSaturateToInt8Grouped(zero)
	return gatherLo64(g.ToBits().ReshapeToUint64s()).ReshapeToUint8s().BitsToInt8()
}

// SaturateToUint8 converts each element of x to uint8, saturating values
// that are outside the range of uint8. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 16 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Int16x16) SaturateToUint8() (z Uint8x32) {
	var zero Int16x16
	g := x.concatSaturateToUint8Grouped(zero)
	return gatherLo64(g.ReshapeToUint64s()).ReshapeToUint8s()
}

// SaturateToUint8 converts each element of x to uint8, saturating values
// that are outside the range of uint8.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 16 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Uint16x16) SaturateToUint8() (z Uint8x32) {
	if X86.AVX512() {
		return x.saturateToUint8()
	}
	// After the Min, x is non-negative as an int16, so the signed-to-unsigned
	// saturation doesn't change it.
	x = x.Min(LoadUint16x16Array(&maxUint8x16))
	return x.BitsToInt16().SaturateToUint8()
}

// SaturateToInt16 converts each element of x to int16, saturating values
// that are outside the range of int16.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Int32x8) SaturateToInt16() (z Int16x16) {
	if X86.AVX512() {
		return x.saturateToInt16()
	}
	var zero Int32x8
	g := x.ConcatSaturateToInt16Grouped(zero)
	return gatherLo64(g.ToBits().ReshapeToUint64s()).ReshapeToUint16s().BitsToInt16()
}

// SaturateToUint16 converts each element of x to uint16, saturating values
// that are outside the range of uint16. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Int32x8) SaturateToUint16() (z Uint16x16) {
	var zero Int32x8
	g := x.ConcatSaturateToUint16Grouped(zero)
	return gatherLo64(g.ReshapeToUint64s()).ReshapeToUint16s()
}

// SaturateToUint16 converts each element of x to uint16, saturating values
// that are outside the range of uint16.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Uint32x8) SaturateToUint16() (z Uint16x16) {
	if X86.AVX512() {
		return x.saturateToUint16()
	}
	// After the Min, x is non-negative as an int32, so the signed-to-unsigned
	// saturation doesn't change it.
	x = x.Min(LoadUint32x8Array(&maxUint16x8))
	return x.BitsToInt32().SaturateToUint16()
}

// SaturateToInt32 converts each element of x to int32, saturating values
// that are outside the range of int32.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Int64x4) SaturateToInt32() (z Int32x8) {
	if X86.AVX512() {
		return x.saturateToInt32()
	}
	hi, lo := LoadInt64x4Array(&maxInt32x4), LoadInt64x4Array(&minInt32x4)
	x = hi.IfElse(x.Greater(hi), x)
	x = lo.IfElse(x.Less(lo), x)
	return x.TruncToInt32()
}

// SaturateToUint32 converts each element of x to uint32, saturating values
// that are outside the range of uint32. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Int64x4) SaturateToUint32() (z Uint32x8) {
	var zero Int64x4
	if X86.AVX512() {
		return x.Max(zero).ToBits().saturateToUint32()
	}
	hi := LoadInt64x4Array(&maxUint32x4)
	x = x.Masked(x.Greater(zero))
	x = hi.IfElse(x.Greater(hi), x)
	return x.ToBits().TruncToUint32()
}

// SaturateToUint32 converts each element of x to uint32, saturating values
// that are outside the range of uint32.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX2
func (x Uint64x4) SaturateToUint32() (z Uint32x8) {
	if X86.AVX512() {
		return x.saturateToUint32()
	}
	// x fits in a uint32 if its high 32 bits are zero.
	var zero Uint64x4
	hi := LoadInt64x4Array(&maxUint32x4).ToBits()
	return x.IfElse(x.ShiftAllRight(32).Equal(zero), hi).TruncToUint32()
}

// SaturateToUint8 converts each element of x to uint8, saturating values
// that are outside the range of uint8. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 32 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int16x32) SaturateToUint8() (z Uint8x64) {
	var zero Int16x32
	return x.Max(zero).ToBits().SaturateToUint8()
}

// SaturateToUint8 converts each element of x to uint8, saturating values
// that are outside the range of uint8. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int32x4) SaturateToUint8() (z Uint8x16) {
	var zero Int32x4
	return x.Max(zero).ToBits().SaturateToUint8()
}

// SaturateToUint8 converts each element of x to uint8, saturating values
// that are outside the range of uint8. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int32x8) SaturateToUint8() (z Uint8x32) {
	var zero Int32x8
	return x.Max(zero).ToBits().SaturateToUint8()
}

// SaturateToUint8 converts each element of x to uint8, saturating values
// that are outside the range of uint8. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 16 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int32x16) SaturateToUint8() (z Uint8x64) {
	var zero Int32x16
	return x.Max(zero).ToBits().SaturateToUint8()
}

// SaturateToUint8 converts each element of x to uint8, saturating values
// that are outside the range of uint8. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 2 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int64x2) SaturateToUint8() (z Uint8x16) {
	var zero Int64x2
	return x.Max(zero).ToBits().SaturateToUint8()
}

// SaturateToUint8 converts each element of x to uint8, saturating values
// that are outside the range of uint8. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int64x4) SaturateToUint8() (z Uint8x32) {
	var zero Int64x4
	return x.Max(zero).ToBits().SaturateToUint8()
}

// SaturateToUint8 converts each element of x to uint8, saturating values
// that are outside the range of uint8. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int64x8) SaturateToUint8() (z Uint8x64) {
	var zero Int64x8
	return x.Max(zero).ToBits().SaturateToUint8()
}

// SaturateToUint16 converts each element of x to uint16, saturating values
// that are outside the range of uint16. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 16 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int32x16) SaturateToUint16() (z Uint16x32) {
	var zero Int32x16
	return x.Max(zero).ToBits().SaturateToUint16()
}

// SaturateToUint16 converts each element of x to uint16, saturating values
// that are outside the range of uint16. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 2 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int64x2) SaturateToUint16() (z Uint16x8) {
	var zero Int64x2
	return x.Max(zero).ToBits().SaturateToUint16()
}

// SaturateToUint16 converts each element of x to uint16, saturating values
// that are outside the range of uint16. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 4 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int64x4) SaturateToUint16() (z Uint16x16) {
	var zero Int64x4
	return x.Max(zero).ToBits().SaturateToUint16()
}

// SaturateToUint16 converts each element of x to uint16, saturating values
// that are outside the range of uint16. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int64x8) SaturateToUint16() (z Uint16x32) {
	var zero Int64x8
	return x.Max(zero).ToBits().SaturateToUint16()
}

// SaturateToUint32 converts each element of x to uint32, saturating values
// that are outside the range of uint32. Negative values become 0.
//
//	z[i] = sat(x[i])
//
// The results are stored in the low 8 elements of z.
// The remaining elements of z are zero.
//
// Emulated, CPU Feature: AVX512
func (x Int64x8) SaturateToUint32() (z Uint32x16) {
	var zero Int64x8
	return x.Max(zero).ToBits().SaturateToUint32()
}