diff --git a/artiq/test/gateware/rtio/__init__.py b/artiq/test/gateware/rtio/__init__.py
new file mode 100644
index 000000000..e69de29bb
diff --git a/artiq/test/gateware/rtio/test_dma.py b/artiq/test/gateware/rtio/test_dma.py
new file mode 100644
index 000000000..5de376afd
--- /dev/null
+++ b/artiq/test/gateware/rtio/test_dma.py
@@ -0,0 +1,69 @@
+import unittest
+
+from migen import *
+from misoc.interconnect import wishbone
+
+from artiq.gateware.rtio import dma
+
+
+def encode_n(n, min_length, max_length):
+    r = []
+    while n:
+        r.append(n & 0xff)
+        n >>= 8
+    r += [0]*(min_length - len(r))
+    if len(r) > max_length:
+        raise ValueError
+    return r
+
+
+def encode_record(channel, timestamp, address, data):
+    r = []
+    r += encode_n(channel, 3, 3)
+    r += encode_n(timestamp, 8, 8)
+    r += encode_n(address, 2, 2)
+    r += encode_n(data, 1, 64)
+    return encode_n(len(r)+1, 1, 1) + r
+
+
+def pack(x, size):
+    r = []
+    for i in range((len(x)+size-1)//size):
+        n = 0
+        for j, w in enumerate(x[i*size:(i+1)*size]):
+            n |= w << j*8
+        r.append(n)
+    return r
+
+
+class TB(Module):
+    def __init__(self, ws):
+        sequence = []
+        sequence += encode_record(0x01, 0x23, 0x12, 0x33)
+        sequence += encode_record(0x901, 0x902, 0x911, 0xeeeeeeeeeeeeeefffffffffffffffffffffffffffffff28888177772736646717738388488)
+        sequence += encode_record(0x81, 0x288, 0x88, 0x8888)
+        sequence.append(0)
+        self.sequence = pack(sequence, ws)
+
+        bus = wishbone.Interface(ws*8)
+        self.submodules.memory = wishbone.SRAM(
+            1024, init=self.sequence, bus=bus)
+        self.submodules.dut = dma.DMA(bus)
+
+        # TODO: remove this hack when misoc supports csr write_from_dev simulation
+        self.sync += If(self.dut.enable.we, self.dut.enable.storage.eq(self.dut.enable.dat_w))
+
+
+class TestDMA(unittest.TestCase):
+    def test_dma(self):
+        ws = 64
+        tb = TB(ws)
+
+        def gen():
+            for i in range(2):
+                yield from tb.dut.enable.write(1)
+                for i in range(30):
+                    print((yield from tb.dut.enable.read()))
+                    yield
+
+        run_simulation(tb, gen(), vcd_name="foo.vcd")