thneed saves 45% of a core (#1512)

* thneed runs the model

* thneed is doing the hooking

* set kernel args

* thneeding the bufferS

* print the images well

* thneeds with better buffers

* includes

* disasm adreno

* parse packets

* disasm works

* disasm better

* more thneeding

* much thneeding

* much more thneeding

* thneed works i think

* thneed is patient

* thneed works

* 7.7%

* gpuobj sync

* yay, it mallocs now

* cleaning it up, Thneed

* sync objs and set power

* thneed needs inputs and outputs

* thneed in modeld

* special modeld runs

* can't thneed the DSP

* test is weird

* thneed modeld uses 6.4% CPU

* add thneed to release

* move to debug

* delete some junk from the pr

* always track the timestamp

* timestamp hacks in thneed

* create a new command queue

* fix timestamp

* pretty much back to what we had, you can't use SNPE with thneed

* improve thneed test

* disable save log

Co-authored-by: Comma Device <device@comma.ai>
old-commit-hash: 302d06ee709cb65dba39e836917f5693f0819efb

selfdrive/modeld/SConscript

@@ -3,6 +3,8 @@ lenv = env.Clone()
 
 libs = [messaging, common, 'OpenCL', 'SNPE', 'capnp', 'zmq', 'kj', 'yuv', gpucommon, visionipc]
 
+TEST_THNEED = False
+
 common_src = [
   "models/commonmodel.c",
   "runners/snpemodel.cc",
@@ -11,6 +13,10 @@ common_src = [
 
 if arch == "aarch64":
   libs += ['gsl', 'CB', 'gnustl_shared']
+  if not TEST_THNEED:
+    common_src += ["thneed/thneed.cc"]
+    lenv['CFLAGS'].append("-DUSE_THNEED")
+    lenv['CXXFLAGS'].append("-DUSE_THNEED")
 elif arch == "larch64":
   libs += ['gsl', 'CB', 'symphony-cpu', 'pthread']
 else:
@@ -34,3 +40,8 @@ lenv.Program('_modeld', [
     "models/driving.cc",
   ]+common, LIBS=libs)
 
+if TEST_THNEED:
+  lenv.Program('thneed/debug/_thneed', [
+      "thneed/thneed.cc", "thneed/debug/test.cc"
+    ]+common, LIBS=libs)
+