},
  "layered_runtime": {
      "layers": [
          {
            "name": "global config",

  },
  "layered_runtime": {
      "layers": [
          {
            "name": "global config",

  },
  "layered_runtime": {
      "layers": [
          {
            "name": "global config",

        "Number of quantized layers with quantized outputs: %d/%d",
        total_quantized_func_count - float_output_func_count,
        total_quantized_func_count));
    lines.push_back(absl::StrFormat("Number of quantize layers added: %d",
                                    quantize_func_count));
    lines.push_back(absl::StrFormat("Number of dequantize layers added: %d",

### Default CA Flow through istio-agent

![CA Flow](docs/ca.svg)

A single SDS request from Envoy goes through a few different layers in istio-agent.

1. First, the request is handled by the SDS server. This is mostly just an intermediate translation layer exposing
   the `SecretManager` to Envoy, without much business logic. For each resource requested by Envoy, the SDS server

  // quantization.
  bool verify_numeric = false;
  // Whether to add verification for layer by layer, or on whole model. When
  // disabled (per-layer) float and quantized ops will be run from same input
  // (output of previous quantized layer). When enabled, float and quantized ops
  // will run with respective float and quantized output of previous ops.
  bool whole_model_verify = false;

      {{- if or .region .zone }}
      ,
      {{- end }}
      "sub_zone": "{{ .sub_zone }}"
      {{- end }}
    },
    "metadata": {{ .meta_json_str }}
  },
  "layered_runtime": {
      "layers": [
          {
            "name": "global config",
            "static_layer": {{ .runtime_flags }}
          },
          {
              "name": "admin",
              "admin_layer": {}
          }
      ]

     * best we can do (since logging may overflow the stack) is to let the error propagate. Because
     * it is an Error, it won't be caught and logged by AbstractFuture.executeListener. Instead, it
     * can propagate through many layers of AbstractTransformFuture up to the root call to set().
     *
     * https://github.com/google/guava/issues/2254
     *
     * Other kinds of Errors are possible:
     *

     * best we can do (since logging may overflow the stack) is to let the error propagate. Because
     * it is an Error, it won't be caught and logged by AbstractFuture.executeListener. Instead, it
     * can propagate through many layers of AbstractTransformFuture up to the root call to set().
     *
     * https://github.com/google/guava/issues/2254
     *
     * Other kinds of Errors are possible:
     *

		// Ideally we would just always build the targets we need but our Makefile is not that smart
		env = append(env, "BUILD_ALL=false")
	}

	env = append(env,
		// Build should already run in container, having multiple layers of docker causes issues
		"BUILD_WITH_CONTAINER=0",
	)
	return env
}

var SkipMake = os.Getenv("SKIP_MAKE")

// RunMake runs a make command for the repo, with standard environment variables set