diff --git a/.travis.yml b/.travis.yml
index 2b759c9..f78749a 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -47,6 +47,7 @@ install:
  - virtualenv /tmp/venv
  - . /tmp/venv/bin/activate
  - pip install pytest
+ - pip install python-subunit
  - grub-mkimage -o ~/grub_x86.efi -O i386-efi normal  echo lsefimmap lsefi lsefisystab efinet tftp minicmd
  - mkdir ~/grub2-arm
  - ( cd ~/grub2-arm; wget -O - http://download.opensuse.org/ports/armv7hl/distribution/leap/42.2/repo/oss/suse/armv7hl/grub2-arm-efi-2.02~beta2-87.1.armv7hl.rpm | rpm2cpio | cpio -di )
diff --git a/test/py/README.md b/test/py/README.md
index aed2fd0..4d9d2b8 100644
--- a/test/py/README.md
+++ b/test/py/README.md
@@ -29,6 +29,7 @@ tests. Similar package names should exist in other distributions.
 | -------------- | ----------------------------- |
 | python         | 2.7.5-5ubuntu3                |
 | python-pytest  | 2.5.1-1                       |
+| python-subunit | -                             |
 | gdisk          | 0.8.8-1ubuntu0.1              |
 | dfu-util       | 0.5-1                         |
 | dtc            | 1.4.0+dfsg-1                  |
diff --git a/tools/binman/binman.py b/tools/binman/binman.py
index f5af535..439908e 100755
--- a/tools/binman/binman.py
+++ b/tools/binman/binman.py
@@ -10,6 +10,7 @@
 """See README for more information"""
 
 import glob
+import multiprocessing
 import os
 import sys
 import traceback
@@ -17,7 +18,7 @@ import unittest
 
 # Bring in the patman and dtoc libraries
 our_path = os.path.dirname(os.path.realpath(__file__))
-for dirname in ['../patman', '../dtoc', '..']:
+for dirname in ['../patman', '../dtoc', '..', '../concurrencytest']:
     sys.path.insert(0, os.path.join(our_path, dirname))
 
 # Bring in the libfdt module
@@ -27,16 +28,22 @@ sys.path.insert(0, os.path.join(our_path,
 
 import cmdline
 import command
+use_concurrent = True
+try:
+    from concurrencytest import ConcurrentTestSuite, fork_for_tests
+except:
+    use_concurrent = False
 import control
 import test_util
 
-def RunTests(debug, args):
+def RunTests(debug, processes, args):
     """Run the functional tests and any embedded doctests
 
     Args:
         debug: True to enable debugging, which shows a full stack trace on error
         args: List of positional args provided to binman. This can hold a test
             name to execute (as in 'binman -t testSections', for example)
+        processes: Number of processes to use to run tests (None=same as #CPUs)
     """
     import elf_test
     import entry_test
@@ -54,19 +61,28 @@ def RunTests(debug, args):
     sys.argv = [sys.argv[0]]
     if debug:
         sys.argv.append('-D')
+    if debug:
+        sys.argv.append('-D')
 
     # Run the entry tests first ,since these need to be the first to import the
     # 'entry' module.
     test_name = args and args[0] or None
+    suite = unittest.TestSuite()
+    loader = unittest.TestLoader()
     for module in (entry_test.TestEntry, ftest.TestFunctional, fdt_test.TestFdt,
                    elf_test.TestElf, image_test.TestImage):
         if test_name:
             try:
-                suite = unittest.TestLoader().loadTestsFromName(test_name, module)
+                suite.addTests(loader.loadTestsFromName(test_name, module))
             except AttributeError:
                 continue
         else:
-            suite = unittest.TestLoader().loadTestsFromTestCase(module)
+            suite.addTests(loader.loadTestsFromTestCase(module))
+    if use_concurrent and processes != 1:
+        concurrent_suite = ConcurrentTestSuite(suite,
+                fork_for_tests(processes or multiprocessing.cpu_count()))
+        concurrent_suite.run(result)
+    else:
         suite.run(result)
 
     print result
@@ -115,7 +131,7 @@ def RunBinman(options, args):
         sys.tracebacklimit = 0
 
     if options.test:
-        ret_code = RunTests(options.debug, args[1:])
+        ret_code = RunTests(options.debug, options.processes, args[1:])
 
     elif options.test_coverage:
         RunTestCoverage()
diff --git a/tools/binman/cmdline.py b/tools/binman/cmdline.py
index f8caa7d..3886d52 100644
--- a/tools/binman/cmdline.py
+++ b/tools/binman/cmdline.py
@@ -46,6 +46,8 @@ def ParseArgs(argv):
     parser.add_option('-p', '--preserve', action='store_true',\
         help='Preserve temporary output directory even if option -O is not '
              'given')
+    parser.add_option('-P', '--processes', type=int,
+                      help='set number of processes to use for running tests')
     parser.add_option('-t', '--test', action='store_true',
                     default=False, help='run tests')
     parser.add_option('-T', '--test-coverage', action='store_true',
diff --git a/tools/binman/entry_test.py b/tools/binman/entry_test.py
index a8bc938..17ab229 100644
--- a/tools/binman/entry_test.py
+++ b/tools/binman/entry_test.py
@@ -13,6 +13,8 @@ import fdt
 import fdt_util
 import tools
 
+entry = None
+
 class TestEntry(unittest.TestCase):
     def setUp(self):
         tools.PrepareOutputDir(None)
@@ -38,7 +40,10 @@ class TestEntry(unittest.TestCase):
     def test2EntryImportLib(self):
         del sys.modules['importlib']
         global entry
-        reload(entry)
+        if entry:
+            reload(entry)
+        else:
+            import entry
         entry.Entry.Create(None, self.GetNode(), 'u-boot-spl')
         del entry
 
diff --git a/tools/binman/ftest.py b/tools/binman/ftest.py
index ed78774..9f21027 100644
--- a/tools/binman/ftest.py
+++ b/tools/binman/ftest.py
@@ -368,6 +368,16 @@ class TestFunctional(unittest.TestCase):
         return pathname
 
     @classmethod
+    def _SetupSplElf(self, src_fname='bss_data'):
+        """Set up an ELF file with a '_dt_ucode_base_size' symbol
+
+        Args:
+            Filename of ELF file to use as SPL
+        """
+        with open(self.TestFile(src_fname)) as fd:
+            TestFunctional._MakeInputFile('spl/u-boot-spl', fd.read())
+
+    @classmethod
     def TestFile(self, fname):
         return os.path.join(self._binman_dir, 'test', fname)
 
@@ -715,8 +725,7 @@ class TestFunctional(unittest.TestCase):
 
     def testImagePadByte(self):
         """Test that the image pad byte can be specified"""
-        with open(self.TestFile('bss_data')) as fd:
-            TestFunctional._MakeInputFile('spl/u-boot-spl', fd.read())
+        self._SetupSplElf()
         data = self._DoReadFile('21_image_pad.dts')
         self.assertEqual(U_BOOT_SPL_DATA + (chr(0xff) * 1) + U_BOOT_DATA, data)
 
@@ -739,6 +748,7 @@ class TestFunctional(unittest.TestCase):
 
     def testPackSorted(self):
         """Test that entries can be sorted"""
+        self._SetupSplElf()
         data = self._DoReadFile('24_sorted.dts')
         self.assertEqual(chr(0) * 1 + U_BOOT_SPL_DATA + chr(0) * 2 +
                          U_BOOT_DATA, data)
@@ -781,6 +791,7 @@ class TestFunctional(unittest.TestCase):
 
     def testPackX86Rom(self):
         """Test that a basic x86 ROM can be created"""
+        self._SetupSplElf()
         data = self._DoReadFile('29_x86-rom.dts')
         self.assertEqual(U_BOOT_DATA + chr(0) * 7 + U_BOOT_SPL_DATA +
                          chr(0) * 2, data)
@@ -1017,15 +1028,13 @@ class TestFunctional(unittest.TestCase):
     def testSplBssPad(self):
         """Test that we can pad SPL's BSS with zeros"""
         # ELF file with a '__bss_size' symbol
-        with open(self.TestFile('bss_data')) as fd:
-            TestFunctional._MakeInputFile('spl/u-boot-spl', fd.read())
+        self._SetupSplElf()
         data = self._DoReadFile('47_spl_bss_pad.dts')
         self.assertEqual(U_BOOT_SPL_DATA + (chr(0) * 10) + U_BOOT_DATA, data)
 
     def testSplBssPadMissing(self):
         """Test that a missing symbol is detected"""
-        with open(self.TestFile('u_boot_ucode_ptr')) as fd:
-            TestFunctional._MakeInputFile('spl/u-boot-spl', fd.read())
+        self._SetupSplElf('u_boot_ucode_ptr')
         with self.assertRaises(ValueError) as e:
             self._DoReadFile('47_spl_bss_pad.dts')
         self.assertIn('Expected __bss_size symbol in spl/u-boot-spl',
@@ -1050,9 +1059,7 @@ class TestFunctional(unittest.TestCase):
             ucode_second: True if the microsecond entry is second instead of
                 third
         """
-        # ELF file with a '_dt_ucode_base_size' symbol
-        with open(self.TestFile('u_boot_ucode_ptr')) as fd:
-            TestFunctional._MakeInputFile('spl/u-boot-spl', fd.read())
+        self._SetupSplElf('u_boot_ucode_ptr')
         first, pos_and_size = self._RunMicrocodeTest(dts, U_BOOT_SPL_NODTB_DATA,
                                                      ucode_second=ucode_second)
         self.assertEqual('splnodtb with microc' + pos_and_size +
@@ -1094,8 +1101,7 @@ class TestFunctional(unittest.TestCase):
         addr = elf.GetSymbolAddress(elf_fname, '__image_copy_start')
         self.assertEqual(syms['_binman_u_boot_spl_prop_offset'].address, addr)
 
-        with open(self.TestFile('u_boot_binman_syms')) as fd:
-            TestFunctional._MakeInputFile('spl/u-boot-spl', fd.read())
+        self._SetupSplElf('u_boot_binman_syms')
         data = self._DoReadFile('53_symbols.dts')
         sym_values = struct.pack('<LQL', 0x24 + 0, 0x24 + 24, 0x24 + 20)
         expected = (sym_values + U_BOOT_SPL_DATA[16:] + chr(0xff) +
@@ -1727,16 +1733,14 @@ class TestFunctional(unittest.TestCase):
 
     def testElf(self):
         """Basic test of ELF entries"""
-        with open(self.TestFile('bss_data')) as fd:
-            TestFunctional._MakeInputFile('spl/u-boot-spl', fd.read())
+        self._SetupSplElf()
         with open(self.TestFile('bss_data')) as fd:
             TestFunctional._MakeInputFile('-boot', fd.read())
         data = self._DoReadFile('96_elf.dts')
 
     def testElfStripg(self):
         """Basic test of ELF entries"""
-        with open(self.TestFile('bss_data')) as fd:
-            TestFunctional._MakeInputFile('spl/u-boot-spl', fd.read())
+        self._SetupSplElf()
         with open(self.TestFile('bss_data')) as fd:
             TestFunctional._MakeInputFile('-boot', fd.read())
         data = self._DoReadFile('97_elf_strip.dts')
diff --git a/tools/concurrencytest/.gitignore b/tools/concurrencytest/.gitignore
new file mode 100644
index 0000000..0d20b64
--- /dev/null
+++ b/tools/concurrencytest/.gitignore
@@ -0,0 +1 @@
+*.pyc
diff --git a/tools/concurrencytest/README.md b/tools/concurrencytest/README.md
new file mode 100644
index 0000000..8e65776
--- /dev/null
+++ b/tools/concurrencytest/README.md
@@ -0,0 +1,74 @@
+concurrencytest
+===============
+
+![testing goats](https://raw.github.com/cgoldberg/concurrencytest/master/testing-goats.png "testing goats")
+
+Python testtools extension for running unittest suites concurrently.
+
+----
+
+Install from PyPI:
+```
+pip install concurrencytest
+```
+
+----
+
+Requires:
+
+ * [testtools](https://pypi.python.org/pypi/testtools) : `pip install testtools`
+ * [python-subunit](https://pypi.python.org/pypi/python-subunit) : `pip install python-subunit`
+
+----
+
+Example:
+
+```python
+import time
+import unittest
+
+from concurrencytest import ConcurrentTestSuite, fork_for_tests
+
+
+class SampleTestCase(unittest.TestCase):
+    """Dummy tests that sleep for demo."""
+
+    def test_me_1(self):
+        time.sleep(0.5)
+
+    def test_me_2(self):
+        time.sleep(0.5)
+
+    def test_me_3(self):
+        time.sleep(0.5)
+
+    def test_me_4(self):
+        time.sleep(0.5)
+
+
+# Load tests from SampleTestCase defined above
+suite = unittest.TestLoader().loadTestsFromTestCase(SampleTestCase)
+runner = unittest.TextTestRunner()
+
+# Run tests sequentially
+runner.run(suite)
+
+# Run same tests across 4 processes
+suite = unittest.TestLoader().loadTestsFromTestCase(SampleTestCase)
+concurrent_suite = ConcurrentTestSuite(suite, fork_for_tests(4))
+runner.run(concurrent_suite)
+```
+Output:
+
+```
+....
+----------------------------------------------------------------------
+Ran 4 tests in 2.003s
+
+OK
+....
+----------------------------------------------------------------------
+Ran 4 tests in 0.504s
+
+OK
+```
diff --git a/tools/concurrencytest/concurrencytest.py b/tools/concurrencytest/concurrencytest.py
new file mode 100644
index 0000000..418d7ee
--- /dev/null
+++ b/tools/concurrencytest/concurrencytest.py
@@ -0,0 +1,144 @@
+#!/usr/bin/env python
+# SPDX-License-Identifier: GPL-2.0+
+#
+# Modified by: Corey Goldberg, 2013
+#
+# Original code from:
+#   Bazaar (bzrlib.tests.__init__.py, v2.6, copied Jun 01 2013)
+#   Copyright (C) 2005-2011 Canonical Ltd
+
+"""Python testtools extension for running unittest suites concurrently.
+
+The `testtools` project provides a ConcurrentTestSuite class, but does
+not provide a `make_tests` implementation needed to use it.
+
+This allows you to parallelize a test run across a configurable number
+of worker processes. While this can speed up CPU-bound test runs, it is
+mainly useful for IO-bound tests that spend most of their time waiting for
+data to arrive from someplace else and can benefit from cocncurrency.
+
+Unix only.
+"""
+
+import os
+import sys
+import traceback
+import unittest
+from itertools import cycle
+from multiprocessing import cpu_count
+
+from subunit import ProtocolTestCase, TestProtocolClient
+from subunit.test_results import AutoTimingTestResultDecorator
+
+from testtools import ConcurrentTestSuite, iterate_tests
+
+
+_all__ = [
+    'ConcurrentTestSuite',
+    'fork_for_tests',
+    'partition_tests',
+]
+
+
+CPU_COUNT = cpu_count()
+
+
+def fork_for_tests(concurrency_num=CPU_COUNT):
+    """Implementation of `make_tests` used to construct `ConcurrentTestSuite`.
+
+    :param concurrency_num: number of processes to use.
+    """
+    def do_fork(suite):
+        """Take suite and start up multiple runners by forking (Unix only).
+
+        :param suite: TestSuite object.
+
+        :return: An iterable of TestCase-like objects which can each have
+        run(result) called on them to feed tests to result.
+        """
+        result = []
+        test_blocks = partition_tests(suite, concurrency_num)
+        # Clear the tests from the original suite so it doesn't keep them alive
+        suite._tests[:] = []
+        for process_tests in test_blocks:
+            process_suite = unittest.TestSuite(process_tests)
+            # Also clear each split list so new suite has only reference
+            process_tests[:] = []
+            c2pread, c2pwrite = os.pipe()
+            pid = os.fork()
+            if pid == 0:
+                try:
+                    stream = os.fdopen(c2pwrite, 'wb', 1)
+                    os.close(c2pread)
+                    # Leave stderr and stdout open so we can see test noise
+                    # Close stdin so that the child goes away if it decides to
+                    # read from stdin (otherwise its a roulette to see what
+                    # child actually gets keystrokes for pdb etc).
+                    sys.stdin.close()
+                    subunit_result = AutoTimingTestResultDecorator(
+                        TestProtocolClient(stream)
+                    )
+                    process_suite.run(subunit_result)
+                except:
+                    # Try and report traceback on stream, but exit with error
+                    # even if stream couldn't be created or something else
+                    # goes wrong.  The traceback is formatted to a string and
+                    # written in one go to avoid interleaving lines from
+                    # multiple failing children.
+                    try:
+                        stream.write(traceback.format_exc())
+                    finally:
+                        os._exit(1)
+                os._exit(0)
+            else:
+                os.close(c2pwrite)
+                stream = os.fdopen(c2pread, 'rb', 1)
+                test = ProtocolTestCase(stream)
+                result.append(test)
+        return result
+    return do_fork
+
+
+def partition_tests(suite, count):
+    """Partition suite into count lists of tests."""
+    # This just assigns tests in a round-robin fashion.  On one hand this
+    # splits up blocks of related tests that might run faster if they shared
+    # resources, but on the other it avoids assigning blocks of slow tests to
+    # just one partition.  So the slowest partition shouldn't be much slower
+    # than the fastest.
+    partitions = [list() for _ in range(count)]
+    tests = iterate_tests(suite)
+    for partition, test in zip(cycle(partitions), tests):
+        partition.append(test)
+    return partitions
+
+
+if __name__ == '__main__':
+    import time
+
+    class SampleTestCase(unittest.TestCase):
+        """Dummy tests that sleep for demo."""
+
+        def test_me_1(self):
+            time.sleep(0.5)
+
+        def test_me_2(self):
+            time.sleep(0.5)
+
+        def test_me_3(self):
+            time.sleep(0.5)
+
+        def test_me_4(self):
+            time.sleep(0.5)
+
+    # Load tests from SampleTestCase defined above
+    suite = unittest.TestLoader().loadTestsFromTestCase(SampleTestCase)
+    runner = unittest.TextTestRunner()
+
+    # Run tests sequentially
+    runner.run(suite)
+
+    # Run same tests across 4 processes
+    suite = unittest.TestLoader().loadTestsFromTestCase(SampleTestCase)
+    concurrent_suite = ConcurrentTestSuite(suite, fork_for_tests(4))
+    runner.run(concurrent_suite)
diff --git a/tools/dtoc/dtoc.py b/tools/dtoc/dtoc.py
index 33b2589..2277af9 100755
--- a/tools/dtoc/dtoc.py
+++ b/tools/dtoc/dtoc.py
@@ -89,6 +89,8 @@ parser.add_option('--include-disabled', action='store_true',
                   help='Include disabled nodes')
 parser.add_option('-o', '--output', action='store', default='-',
                   help='Select output filename')
+parser.add_option('-P', '--processes', type=int,
+                  help='set number of processes to use for running tests')
 parser.add_option('-t', '--test', action='store_true', dest='test',
                   default=False, help='run tests')
 parser.add_option('-T', '--test-coverage', action='store_true',
diff --git a/tools/dtoc/test_fdt.py b/tools/dtoc/test_fdt.py
index 2e6febe..8d70dd2 100755
--- a/tools/dtoc/test_fdt.py
+++ b/tools/dtoc/test_fdt.py
@@ -547,6 +547,8 @@ if __name__ != '__main__':
 parser = OptionParser()
 parser.add_option('-B', '--build-dir', type='string', default='b',
         help='Directory containing the build output')
+parser.add_option('-P', '--processes', type=int,
+                  help='set number of processes to use for running tests')
 parser.add_option('-t', '--test', action='store_true', dest='test',
                   default=False, help='run tests')
 parser.add_option('-T', '--test-coverage', action='store_true',
diff --git a/tools/patman/test_util.py b/tools/patman/test_util.py
index 0e79af8..687d407 100644
--- a/tools/patman/test_util.py
+++ b/tools/patman/test_util.py
@@ -43,7 +43,7 @@ def RunTestCoverage(prog, filter_fname, exclude_list, build_dir, required=None):
     glob_list += exclude_list
     glob_list += ['*libfdt.py', '*site-packages*']
     cmd = ('PYTHONPATH=$PYTHONPATH:%s/sandbox_spl/tools python-coverage run '
-           '--omit "%s" %s -t' % (build_dir, ','.join(glob_list), prog))
+           '--omit "%s" %s -P1 -t' % (build_dir, ','.join(glob_list), prog))
     os.system(cmd)
     stdout = command.Output('python-coverage', 'report')
     lines = stdout.splitlines()