Use proper queue in HostReceiveNode for speed

This makes things faster by not requiring us to increment a counter
through the node, and also makes sure we get all the data.

We also fix the time in this node to use zero-based timesteps
rather than one-based timesteps (to conform with core Nengo).

nengo_loihi/builder/inputs.py

@@ -1,3 +1,5 @@
+import collections
+
 from nengo import Node
 from nengo.exceptions import SimulationError
 from nengo.params import Default
@@ -24,21 +26,20 @@ class HostReceiveNode(Node):
     """For receiving chip->host messages"""
 
     def __init__(self, dimensions, label=Default, check_output=False):
-        self.queue = [(0, np.zeros(dimensions))]
-        self.queue_index = 0
+        self.queue = collections.deque()
         super(HostReceiveNode, self).__init__(
             self.update, size_in=0, size_out=dimensions, label=label
         )
         if hasattr(self, "check_output"):
             self.check_output = check_output
 
     def update(self, t):
-        while (
-            len(self.queue) > self.queue_index + 1
-            and self.queue[self.queue_index][0] < t
-        ):
-            self.queue_index += 1
-        return self.queue[self.queue_index][1]
+        if t <= 0:
+            return np.zeros(self.size_out)
+
+        t1, x = self.queue.popleft()
+        assert abs(t - t1) < 1e-8
+        return x
 
     def receive(self, t, x):
         self.queue.append((t, x))

nengo_loihi/emulator/interface.py

@@ -638,7 +638,7 @@ def send(self, probe, already_sent, receiver):
             if probe.weights is not None:
                 x = np.dot(x, probe.weights)
             for j, xx in enumerate(x):
-                receiver.receive(self.dt * (already_sent + j + 2), xx)
+                receiver.receive(self.dt * (already_sent + j + 1), xx)
         return len(x)
 
     def update(self, t, compartment):

nengo_loihi/hardware/interface.py

@@ -202,7 +202,7 @@ def _chip2host_monitor(self, probes_receivers):
 
                 for j in range(len(x)):
                     receiver.receive(
-                        self.model.dt * (self._chip2host_sent_steps + j + 2), x[j]
+                        self.model.dt * (self._chip2host_sent_steps + j + 1), x[j]
                     )
 
         if increment is not None:

nengo_loihi/hardware/snips/nengo_io.c.template

@@ -173,7 +173,7 @@ void nengo_io(runState *s) {
         }
     }
 
-    buffer[0] = s->{{ obfs.step }};
+    buffer[0] = s->{{ obfs.step }} - 1;  // zero-based time step
 {% for n_out, core, compartment, key in probes %}
 {% if key == 'u' %}
     buffer[{{ n_out }}] = core{{ core }}->{{ obfs.state }}[{{ compartment }}].U;