* add thneed optimizer
* local work group opt
* kernels and final mods
* release files
* build system touchups
* fix kernel path, rand inputs for self test
* broken since extra is gone
* update model replay ref
Co-authored-by: Comma Device <device@comma.ai>
* completely untested
* it builds now
* bug fixes, save 1ms
* using a kernel to copy works
* more sane API to loadyuv
Co-authored-by: Comma Device <device@comma.ai>
* use cstring instead of string.h
* use cstdio instead of stdio.h
* remove inttypes.h
* use cstdlib instead of stdlib.h
* use cstdint instead of stdint.h
* #include <cstddef>
* cstdlib
* use cmath
* remove stddef.h
* use cassert
* use csignal
* use ctime
* use cerror
* rebase master
* start thneed load/save
* compiling
* fix loading
* build thneed model in scons
* don't hardcode /data/openpilot
* release files
* those too
* support for loading/saving binary kernels
* save binaries out of json band
* make binary a command line flag to the compiler
* need include assert
* fix shadowed common in SConscript
* cleanup run.h
* hmm, the recurrent buffer wasn't 0ed
* ugh, unique ptr
* remove power constraint, refactor record
* Revert "remove power constraint, refactor record"
This reverts commit bb6fa52db6df59cd9d6420a6f630430e35af8a5e.
* print on thneed stop
* fingers crossed for this one
* recorded
* just curious
* okay okay, pass tests?
* cleanups
* refactor wait
Co-authored-by: Comma Device <device@comma.ai>
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
* bigmodel
* more debug print
* debugging bigmodel
* remove the tanh, debugging
* print images/buffers
* disassemble the command queues
* decompiler
* dump the shaders
* full disasm
* support patching kernel and fixing convolution_horizontal_reduced_reads_1x1
* microbenchmark
* 42 GFLOPS, 1 GB/s
* gemm benchmark
* 75 GFLOPS vs 42 GFLOPS
* 115 GFLOPS
* oops, never mind
* gemm image is slow
* this is pretty hopeless
* gemm image gets 62 GFLOPS
* this is addictive and still a waste of time
* cleanup cleanup
* that hook was dumb
* tabbing
* more tabbing
Co-authored-by: Comma Device <device@comma.ai>
* 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>
George Hotz
Dean Lee