eval in arrow function

var testCase = function (actual, expected, message) {
  if (actual !== expected) {
    throw message + ". Expected '" + expected + "', but was '" + actual + "'";
  }
};

 var testValue  = 'test-value';

 var A = class A {
    constructor() {
      this.idValue = testValue;
    }
};

 var B = class B extends A {
  constructor (beforeSuper) {
    var arrow = () => eval('(() => this.idValue)()');

     if (beforeSuper) {
      var result = arrow();
      super();
      testCase(result, testValue, "Error: has to be TDZ error");
    } else {
      super();
      let result= arrow();
      testCase(result, testValue, "Error: super() should create this and put value into idValue property");
    }
  }
};

 for (var i=0; i < 10000; i++) {
   var b = new B(false);
}

// Fixme: Failed test with 'Segmentation fault: 11' error in release mode in 6 cases from 14. List of failed case:
No, it doesn't. I think it should be fixed in separate patch, because it looks like some bug in optimization, and I don't have enough skill to fix it because it appear in only in release mode and only in listed type of tests. 
// dfg-maximal-flush-validate-no-cjit, ftl-no-cjit-no-inline-validate, ftl-no-cjit-no-put-stack-validate, 
// ftl-no-cjit-small-pool, ftl-no-cjit-validate, no-cjit-validate-phases
/*
var testException = function (value1, index) {
  var exception;
  try {
       new B(value1);
  } catch (e) {
      exception = e;
      if (!(e instanceof ReferenceError))
          throw "Exception thrown was not a reference error";
  }

   if (!exception)
      throw "Exception not thrown for an unitialized this at iteration #" + index;
}


 for (var i=0; i < 10000; i++) {
   testException(true, i);
}/* */