}], [{
      TypeAttr::get(getResult().getType().cast<TensorType>().getElementType())
    }]>;
}

def TFL_CeilOp: TFL_Op<"ceil", [
    Pure,
    TF_SameOperandsAndResultTypeResolveRef]> {
  let summary = "Ceil operator";

  let description = [{
    Returns element-wise ceil value of the input.
  }];

  let arguments = (ins TFL_FpTensor:$x);

  let results = (outs TFL_FpTensor:$y);

		} else if policy.Type == autoscalingv2.PercentScalingPolicy {
			// the proposal has to be rounded up because the proposed change might not increase the replica count causing the target to never scale up
			proposed = int32(math.Ceil(float64(periodStartReplicas) * (1 + float64(policy.Value)/100)))
		}
		result = selectPolicyFn(result, proposed)
	}
	return result
}

	clientConfig.DynamicRecordSizingDisabled = dynamicRecordSizingDisabled
	clientConfig.MaxVersion = version

	buf := make([]byte, bufsize)
	chunks := int(math.Ceil(float64(totalBytes) / float64(len(buf))))
	for i := 0; i < N; i++ {
		conn, err := Dial("tcp", ln.Addr().String(), clientConfig)
		if err != nil {
			b.Fatal(err)
		}
		for j := 0; j < chunks; j++ {

      new absl::flat_hash_map<string, std::vector<string>>{
          // Unary
          {"PW",
           {"ComplexAbs", "Angle", "Conj", "Abs", "Acos", "Acosh", "Asin",
            "Atan", "Atanh", "Ceil", "Cos", "Cosh", "Sin", "Exp", "Expm1",
            "Floor", "IsFinite", "IsInf", "IsNan", "Inv", "Reciprocal", "Log",
            "Log1p", "Invert", "LogicalNot", "Ndtri", "Neg", "Rint", "Round",

(Select1 (Div128u (Const64 [0]) lo y)) => (Mod64u lo y)

(Not (ConstBool [c])) => (ConstBool [!c])

(Floor       (Const64F [c])) => (Const64F [math.Floor(c)])
(Ceil        (Const64F [c])) => (Const64F [math.Ceil(c)])
(Trunc       (Const64F [c])) => (Const64F [math.Trunc(c)])
(RoundToEven (Const64F [c])) => (Const64F [math.RoundToEven(c)])

// Convert x * 1 to x.

            "format": "int32",...

            "format": "int32",...

  // tensorflow/lite/kernels/range.cc.
  FloatOrInt diff = limit - start;
  if (std::is_integral<FloatOrInt>::value) {
    return ((std::abs(diff) + std::abs(delta) - 1) / std::abs(delta));
  }
  return std::ceil(std::abs(diff / delta));
}

// Builds a constant range tensor of `result_elem_type` elements.
// Template parameter `FloatOrIntAtrr` must be mlir::IntegerAttr or
// mlir::FloatAttr.
template <typename FloatOrIntAtrr>

(PopCount8 x) => (POPCNTL (MOVBQZX <typ.UInt32> x))

(Sqrt ...) => (SQRTSD ...)
(Sqrt32 ...) => (SQRTSS ...)

(RoundToEven x) => (ROUNDSD [0] x)
(Floor x)       => (ROUNDSD [1] x)
(Ceil x)        => (ROUNDSD [2] x)
(Trunc x)       => (ROUNDSD [3] x)

(FMA x y z) => (VFMADD231SD z x y)

// Lowering extension
// Note: we always extend to 64 bits even though some ops don't need that many result bits.

	resourcesUsedRatio := float64(totalUsedCPUOfAllPods) / float64(float64(totalRequestedCPUOfAllPods)*target)

	// i.e. .60 * 20 -> scaled down expectation.
	finalPods := int32(math.Ceil(resourcesUsedRatio * float64(startPods)))

	// To breach tolerance we will create a utilization ratio difference of tolerance to usageRatioToleranceValue)
	tc1 := testCase{
		minReplicas:             0,