diff --git a/meta/lib/oeqa/oetest.py b/meta/lib/oeqa/oetest.py
new file mode 100644
index 0000000..0fe68d4
--- /dev/null
+++ b/meta/lib/oeqa/oetest.py
@@ -0,0 +1,214 @@
+# Copyright (C) 2013 Intel Corporation
+#
+# Released under the MIT license (see COPYING.MIT)
+
+# Main unittest module used by testimage.bbclass
+# This provides the oeRuntimeTest base class which is inherited by all tests in meta/lib/oeqa/runtime.
+
+# It also has some helper functions and it's responsible for actually starting the tests
+
+import os, re, mmap
+import unittest
+import inspect
+import subprocess
+import bb
+from oeqa.utils.decorators import LogResults, gettag
+from sys import exc_info, exc_clear
+
+def getVar(obj):
+    #extend form dict, if a variable didn't exists, need find it in testcase
+    class VarDict(dict):
+        def __getitem__(self, key):
+            return gettag(obj, key)
+    return VarDict()
+
+def checkTags(tc, tagexp):
+    return eval(tagexp, None, getVar(tc))
+
+
+def filterByTagExp(testsuite, tagexp):
+    if not tagexp:
+        return testsuite
+    caseList = []
+    for each in testsuite:
+        if not isinstance(each, unittest.BaseTestSuite):
+            if checkTags(each, tagexp):
+                caseList.append(each)
+        else:
+            caseList.append(filterByTagExp(each, tagexp))
+    return testsuite.__class__(caseList)
+
+def loadTests(tc, type="runtime"):
+    if type == "runtime":
+        # set the context object passed from the test class
+        setattr(oeTest, "tc", tc)
+        # set ps command to use
+        setattr(oeRuntimeTest, "pscmd", "ps -ef" if oeTest.hasPackage("procps") else "ps")
+        # prepare test suite, loader and runner
+        suite = unittest.TestSuite()
+    elif type == "sdk":
+        # set the context object passed from the test class
+        setattr(oeTest, "tc", tc)
+    testloader = unittest.TestLoader()
+    testloader.sortTestMethodsUsing = None
+    suites = [testloader.loadTestsFromName(name) for name in tc.testslist]
+    suites = filterByTagExp(suites, getattr(tc, "tagexp", None))
+
+    def getTests(test):
+        '''Return all individual tests executed when running the suite.'''
+        # Unfortunately unittest does not have an API for this, so we have
+        # to rely on implementation details. This only needs to work
+        # for TestSuite containing TestCase.
+        method = getattr(test, '_testMethodName', None)
+        if method:
+            # leaf case: a TestCase
+            yield test
+        else:
+            # Look into TestSuite.
+            tests = getattr(test, '_tests', [])
+            for t1 in tests:
+                for t2 in getTests(t1):
+                    yield t2
+
+    # Determine dependencies between suites by looking for @skipUnlessPassed
+    # method annotations. Suite A depends on suite B if any method in A
+    # depends on a method on B.
+    for suite in suites:
+        suite.dependencies = []
+        suite.depth = 0
+        for test in getTests(suite):
+            methodname = getattr(test, '_testMethodName', None)
+            if methodname:
+                method = getattr(test, methodname)
+                depends_on = getattr(method, '_depends_on', None)
+                if depends_on:
+                    for dep_suite in suites:
+                        if depends_on in [getattr(t, '_testMethodName', None) for t in getTests(dep_suite)]:
+                            if dep_suite not in suite.dependencies and \
+                               dep_suite is not suite:
+                                suite.dependencies.append(dep_suite)
+                            break
+                    else:
+                        bb.warn("Test %s was declared as @skipUnlessPassed('%s') but that test is either not defined or not active. Will run the test anyway." %
+                                (test, depends_on))
+    # Use brute-force topological sort to determine ordering. Sort by
+    # depth (higher depth = must run later), with original ordering to
+    # break ties.
+    def set_suite_depth(suite):
+        for dep in suite.dependencies:
+            new_depth = set_suite_depth(dep) + 1
+            if new_depth > suite.depth:
+                suite.depth = new_depth
+        return suite.depth
+    for index, suite in enumerate(suites):
+        set_suite_depth(suite)
+        suite.index = index
+    suites.sort(cmp=lambda a,b: cmp((a.depth, a.index), (b.depth, b.index)))
+    return testloader.suiteClass(suites)
+
+def runTests(tc, type="runtime"):
+
+    suite = loadTests(tc, type)
+    bb.note("Test modules  %s" % tc.testslist)
+    if hasattr(tc, "tagexp") and tc.tagexp:
+        bb.note("Filter test cases by tags: %s" % tc.tagexp)
+    bb.note("Found %s tests" % suite.countTestCases())
+    runner = unittest.TextTestRunner(verbosity=2)
+    result = runner.run(suite)
+
+    return result
+
+@LogResults
+class oeTest(unittest.TestCase):
+
+    longMessage = True
+
+    @classmethod
+    def hasPackage(self, pkg):
+        for item in oeTest.tc.pkgmanifest.split('\n'):
+            if re.match(pkg, item):
+                return True
+        return False
+
+    @classmethod
+    def hasFeature(self,feature):
+
+        if feature in oeTest.tc.imagefeatures or \
+                feature in oeTest.tc.distrofeatures:
+            return True
+        else:
+            return False
+
+class oeRuntimeTest(oeTest):
+    def __init__(self, methodName='runTest'):
+        self.target = oeRuntimeTest.tc.target
+        super(oeRuntimeTest, self).__init__(methodName)
+
+    def setUp(self):
+        # Check if test needs to run
+        if self.tc.sigterm:
+            self.fail("Got SIGTERM")
+        elif (type(self.target).__name__ == "QemuTarget"):
+            self.assertTrue(self.target.check(), msg = "Qemu not running?")
+
+    def tearDown(self):
+        # If a test fails or there is an exception
+        if not exc_info() == (None, None, None):
+            exc_clear()
+            #Only dump for QemuTarget
+            if (type(self.target).__name__ == "QemuTarget"):
+                self.tc.host_dumper.create_dir(self._testMethodName)
+                self.tc.host_dumper.dump_host()
+                self.target.target_dumper.dump_target(
+                        self.tc.host_dumper.dump_dir)
+                print ("%s dump data stored in %s" % (self._testMethodName,
+                         self.tc.host_dumper.dump_dir))
+
+    #TODO: use package_manager.py to install packages on any type of image
+    def install_packages(self, packagelist):
+        for package in packagelist:
+            (status, result) = self.target.run("smart install -y "+package)
+            if status != 0:
+                return status
+
+class oeSDKTest(oeTest):
+    def __init__(self, methodName='runTest'):
+        self.sdktestdir = oeSDKTest.tc.sdktestdir
+        super(oeSDKTest, self).__init__(methodName)
+
+    @classmethod
+    def hasHostPackage(self, pkg):
+
+        if re.search(pkg, oeTest.tc.hostpkgmanifest):
+            return True
+        return False
+
+    def _run(self, cmd):
+        return subprocess.check_output(cmd, shell=True)
+
+def getmodule(pos=2):
+    # stack returns a list of tuples containg frame information
+    # First element of the list the is current frame, caller is 1
+    frameinfo = inspect.stack()[pos]
+    modname = inspect.getmodulename(frameinfo[1])
+    #modname = inspect.getmodule(frameinfo[0]).__name__
+    return modname
+
+def skipModule(reason, pos=2):
+    modname = getmodule(pos)
+    if modname not in oeTest.tc.testsrequired:
+        raise unittest.SkipTest("%s: %s" % (modname, reason))
+    else:
+        raise Exception("\nTest %s wants to be skipped.\nReason is: %s" \
+                "\nTest was required in TEST_SUITES, so either the condition for skipping is wrong" \
+                "\nor the image really doesn't have the required feature/package when it should." % (modname, reason))
+
+def skipModuleIf(cond, reason):
+
+    if cond:
+        skipModule(reason, 3)
+
+def skipModuleUnless(cond, reason):
+
+    if not cond:
+        skipModule(reason, 3)