Compressed vipc improvements (#28837)

* debug flag

* make it easier to use
old-commit-hash: 9774fb43e9

tools/camerastream/compressed_vipc.py

@@ -13,8 +13,16 @@ from cereal.visionipc import VisionIpcServer, VisionStreamType
 W, H = 1928, 1208
 V4L2_BUF_FLAG_KEYFRAME = 8
 
-def decoder(addr, sock_name, vipc_server, vst, nvidia):
+ENCODE_SOCKETS = {
-  print("start decoder for %s" % sock_name)
+  VisionStreamType.VISION_STREAM_ROAD: "roadEncodeData",
+  VisionStreamType.VISION_STREAM_WIDE_ROAD: "wideRoadEncodeData",
+  VisionStreamType.VISION_STREAM_DRIVER: "driverEncodeData",
+}
+
+def decoder(addr, vipc_server, vst, nvidia, debug=False):
+  sock_name = ENCODE_SOCKETS[vst]
+  if debug:
+    print("start decoder for %s" % sock_name)
   if nvidia:
     os.environ["NV_LOW_LATENCY"] = "3"    # both bLowLatency and CUVID_PKT_ENDOFPICTURE
     sys.path += os.environ["LD_LIBRARY_PATH"].split(":")
@@ -40,11 +48,12 @@ def decoder(addr, sock_name, vipc_server, vst, nvidia):
     msgs = messaging.drain_sock(sock, wait_for_one=True)
     for evt in msgs:
       evta = getattr(evt, evt.which())
-      if evta.idx.encodeId != 0 and evta.idx.encodeId != (last_idx+1):
+      if debug and evta.idx.encodeId != 0 and evta.idx.encodeId != (last_idx+1):
         print("DROP PACKET!")
       last_idx = evta.idx.encodeId
       if not seen_iframe and not (evta.idx.flags & V4L2_BUF_FLAG_KEYFRAME):
-        print("waiting for iframe")
+        if debug:
+          print("waiting for iframe")
         continue
       time_q.append(time.monotonic())
       network_latency = (int(time.time()*1e9) - evta.unixTimestampNanos)/1e6
@@ -62,14 +71,16 @@ def decoder(addr, sock_name, vipc_server, vst, nvidia):
       if nvidia:
         rawSurface = nvDec.DecodeSurfaceFromPacket(np.frombuffer(evta.data, dtype=np.uint8))
         if rawSurface.Empty():
-          print("DROP SURFACE")
+          if debug:
+            print("DROP SURFACE")
           continue
         convSurface = conv_yuv.Execute(rawSurface, cc1)
         nvDwn_yuv.DownloadSingleSurface(convSurface, img_yuv)
       else:
         frames = codec.decode(av.packet.Packet(evta.data))
         if len(frames) == 0:
-          print("DROP SURFACE")
+          if debug:
+            print("DROP SURFACE")
           continue
         assert len(frames) == 1
         img_yuv = frames[0].to_ndarray(format=av.video.format.VideoFormat('yuv420p')).flatten()
@@ -83,34 +94,45 @@ def decoder(addr, sock_name, vipc_server, vst, nvidia):
 
       pc_latency = (time.monotonic()-time_q[0])*1000
       time_q = time_q[1:]
-      print("%2d %4d %.3f %.3f roll %6.2f ms latency %6.2f ms + %6.2f ms + %6.2f ms = %6.2f ms" % (len(msgs), evta.idx.encodeId, evt.logMonoTime/1e9, evta.idx.timestampEof/1e6, frame_latency, process_latency, network_latency, pc_latency, process_latency+network_latency+pc_latency ), len(evta.data), sock_name)
+      if debug:
+        print("%2d %4d %.3f %.3f roll %6.2f ms latency %6.2f ms + %6.2f ms + %6.2f ms = %6.2f ms" % (len(msgs), evta.idx.encodeId, evt.logMonoTime/1e9, evta.idx.timestampEof/1e6, frame_latency, process_latency, network_latency, pc_latency, process_latency+network_latency+pc_latency ), len(evta.data), sock_name)
+
+class CompressedVipc:
+  def __init__(self, addr, vision_streams, nvidia=False, debug=False):
+    self.vipc_server = VisionIpcServer("camerad")
+    for vst in vision_streams:
+      self.vipc_server.create_buffers(vst, 4, False, W, H)
+    self.vipc_server.start_listener()
 
-def main(addr, cams, nvidia=False):
+    self.procs = []
-  vipc_server = VisionIpcServer("camerad")
+    for vst in vision_streams:
-  for vst in cams.values():
+      p = multiprocessing.Process(target=decoder, args=(addr, self.vipc_server, vst, nvidia, debug))
-    vipc_server.create_buffers(vst, 4, False, W, H)
+      p.start()
-  vipc_server.start_listener()
+      self.procs.append(p)
 
-  procs = []
+  def join(self):
-  for k, v in cams.items():
+    for p in self.procs:
-    p = multiprocessing.Process(target=decoder, args=(addr, k, vipc_server, v, nvidia))
+      p.join()
-    p.start()
-    procs.append(p)
 
-  for p in procs:
+  def kill(self):
-    p.join()
+    for p in self.procs:
+      p.terminate()
+    self.join()
 
 if __name__ == "__main__":
   parser = argparse.ArgumentParser(description="Decode video streams and broadcast on VisionIPC")
   parser.add_argument("addr", help="Address of comma three")
   parser.add_argument("--nvidia", action="store_true", help="Use nvidia instead of ffmpeg")
   parser.add_argument("--cams", default="0,1,2", help="Cameras to decode")
+  parser.add_argument("--silent", action="store_true", help="Suppress debug output")
   args = parser.parse_args()
 
-  all_cams = [
+  vision_streams = [
-    ("roadEncodeData", VisionStreamType.VISION_STREAM_ROAD),
+    VisionStreamType.VISION_STREAM_ROAD,
-    ("wideRoadEncodeData", VisionStreamType.VISION_STREAM_WIDE_ROAD),
+    VisionStreamType.VISION_STREAM_WIDE_ROAD,
-    ("driverEncodeData", VisionStreamType.VISION_STREAM_DRIVER),
+    VisionStreamType.VISION_STREAM_DRIVER,
   ]
-  cams = dict([all_cams[int(x)] for x in args.cams.split(",")])
+
-  main(args.addr, cams, args.nvidia)
+  vsts = [vision_streams[int(x)] for x in args.cams.split(",")]
+  cvipc = CompressedVipc(args.addr, vsts, args.nvidia, debug=(not args.silent))
+  cvipc.join()