Successfully registered all kernels from shared library: NOT_SUPPORTED

[ismukhin]

🐛 Describe the bug

#include <iostream>
#include <cassert>
#include <executorch/extension/module/module.h>
#include <executorch/extension/tensor/tensor.h>

using namespace ::executorch::extension;

int main() {
  Module module("mv2_xnnpack.pte");
  const auto error = module.load();

  assert(module.is_loaded());
  float input[1 * 3 * 224 * 224];
  auto tensor = from_blob(input, {1, 3, 224, 224});
  const auto result = module.execute("forward", tensor);
  if (result.ok()) {
    // Retrieve the output data.
    std::cout << "HERE" << std::endl;
    const auto output = result->at(0).toTensor().const_data_ptr<float>();
  }
}

Build with CMake:

cmake_minimum_required(VERSION 3.21)

project(exec)

set(CMAKE_CXX_STANDARD 17)
set(CMAKE_CXX_STANDARD_REQUIRED ON)
set(CMAKE_CXX_EXTENSIONS OFF)


set(executorch_DIR "executorch/build1/executorch_install/lib/cmake/ExecuTorch")

find_package(executorch REQUIRED)

set(_common_compile_options -Wno-deprecated-declarations)

add_executable(exec main.cpp)

include_directories("${CMAKE_CURRENT_SOURCE_DIR}")

target_link_libraries(
    exec PRIVATE
  ${EXECUTORCH_LIBRARIES}
)

target_link_options(exec PRIVATE -Wl,--whole-archive,--allow-multiple-definition)

message(STATUS ${EXECUTORCH_LIBRARIES})
message(STATUS ${EXECUTORCH_INCLUDE_DIRS})
message(STATUS ${libs})
message(STATUS ${EXECUTORCH_LIBRARY})

Executorch build command:

cmake -DEXECUTORCH_BUILD_EXTENSION_DATA_LOADER=ON \ 
-DEXECUTORCH_BUILD_EXTENSION_MODULE=ON \
-DEXECUTORCH_BUILD_EXTENSION_RUNNER_UTIL=ON \
-DEXECUTORCH_BUILD_EXTENSION_TENSOR=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_DEVTOOLS=ON \
-DBUILD_EXECUTORCH_PORTABLE_OPS=ON \
-DEXECUTORCH_BUILD_PYBIND=ON \
-DEXECUTORCH_BUILD_XNNPACK=ON ../ && \
cmake --build . -j$(nproc --all) && \

cmake --install . --prefix executorch_install

Get the error:

D 00:00:00.000184 executorch:operator_registry.cpp:92] Successfully registered all kernels from shared library: NOT_SUPPORTED
E 00:00:00.000222 executorch:operator_registry.cpp:85] Re-registering aten::sym_size.int, from NOT_SUPPORTED
E 00:00:00.000232 executorch:operator_registry.cpp:86] key: (null), is_fallback: true
F 00:00:00.000233 executorch:operator_registry.cpp:106] In function register_kernels(), assert failed (false): Kernel registration failed with error 18, see error log for details.
Aborted (core dumped)

Code of model export:

import torch
import torchvision.models as models
from torchvision.models.mobilenetv2 import MobileNet_V2_Weights
from executorch.backends.xnnpack.partition.xnnpack_partitioner import XnnpackPartitioner
from executorch.exir import to_edge_transform_and_lower

mobilenet_v2 = models.mobilenetv2.mobilenet_v2(weights=MobileNet_V2_Weights.DEFAULT).eval()
sample_inputs = (torch.randn(1, 3, 224, 224), )

et_program = to_edge_transform_and_lower(
    torch.export.export(mobilenet_v2, sample_inputs),
    #partitioner=[XnnpackPartitioner()],
).to_executorch()

with open("mv2_xnnpack.pte", "wb") as file:
    et_program.write_to_file(file)


from pathlib import Path
from executorch.runtime import Verification, Runtime, Program, Method

et_runtime: Runtime = Runtime.get()
program: Program = et_runtime.load_program(
    Path("mv2_xnnpack.pte"),
    verification=Verification.Minimal,
)
print("Program methods:", program.method_names)
forward: Method = program.load_method("forward")

inputs = (torch.ones(1, 3, 224, 224),)
outputs = forward.execute(inputs)
print(f"Ran forward({inputs})")
print(f"  outputs: {outputs}")

Versions

Collecting environment information...
PyTorch version: 2.6.0+cpu
Is debug build: False
CUDA used to build PyTorch: Could not collect
ROCM used to build PyTorch: N/A

OS: Ubuntu 24.04.2 LTS (x86_64)
GCC version: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0
Clang version: 18.1.3 (1ubuntu1)
CMake version: version 3.31.6
Libc version: glibc-2.39

Python version: 3.10.0 (default, Mar  3 2022, 09:58:08) [GCC 7.5.0] (64-bit runtime)
Python platform: Linux-5.15.167.4-microsoft-standard-WSL2-x86_64-with-glibc2.39
Is CUDA available: False
CUDA runtime version: Could not collect
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: GPU 0: NVIDIA GeForce RTX 4070 SUPER
Nvidia driver version: 572.42
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture:                         x86_64
CPU op-mode(s):                       32-bit, 64-bit
Address sizes:                        39 bits physical, 48 bits virtual
Byte Order:                           Little Endian
CPU(s):                               20
On-line CPU(s) list:                  0-19
Vendor ID:                            GenuineIntel
Model name:                           Intel(R) Core(TM) i5-14600KF
CPU family:                           6
Model:                                183
Thread(s) per core:                   2
Core(s) per socket:                   10
Socket(s):                            1
Stepping:                             1
BogoMIPS:                             6988.79
Flags:                                fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology tsc_reliable nonstop_tsc cpuid pni pclmulqdq vmx ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves avx_vnni umip waitpkg gfni vaes vpclmulqdq rdpid movdiri movdir64b fsrm md_clear serialize flush_l1d arch_capabilities
Virtualization:                       VT-x
Hypervisor vendor:                    Microsoft
Virtualization type:                  full
L1d cache:                            480 KiB (10 instances)
L1i cache:                            320 KiB (10 instances)
L2 cache:                             20 MiB (10 instances)
L3 cache:                             24 MiB (1 instance)
Vulnerability Gather data sampling:   Not affected
Vulnerability Itlb multihit:          Not affected
Vulnerability L1tf:                   Not affected
Vulnerability Mds:                    Not affected
Vulnerability Meltdown:               Not affected
Vulnerability Mmio stale data:        Not affected
Vulnerability Reg file data sampling: Mitigation; Clear Register File
Vulnerability Retbleed:               Mitigation; Enhanced IBRS
Vulnerability Spec rstack overflow:   Not affected
Vulnerability Spec store bypass:      Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:             Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:             Mitigation; Enhanced / Automatic IBRS; IBPB conditional; RSB filling; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S
Vulnerability Srbds:                  Not affected
Vulnerability Tsx async abort:        Not affected

Versions of relevant libraries:
[pip3] executorch==0.5.0a0+1bc0699
[pip3] numpy==2.0.0
[pip3] torch==2.6.0+cpu
[pip3] torchao==0.8.0+gitebc43034
[pip3] torchaudio==2.6.0+cpu
[pip3] torchsr==1.0.4
[pip3] torchvision==0.21.0+cpu
[conda] executorch                0.5.0a0+1bc0699          pypi_0    pypi
[conda] numpy                     2.0.0                    pypi_0    pypi
[conda] torch                     2.6.0+cpu                pypi_0    pypi
[conda] torchao                   0.8.0+gitebc43034          pypi_0    pypi
[conda] torchaudio                2.6.0+cpu                pypi_0    pypi
[conda] torchsr                   1.0.4                    pypi_0    pypi
[conda] torchvision               0.21.0+cpu               pypi_0    pypi

I try to link builded executorch to my project with find_package, but get the folowing error when try start inference. But if I use add_subdirectory(executorch) like:

cmake_minimum_required(VERSION 3.21)

project(exec)

set(CMAKE_CXX_STANDARD 17)
set(CMAKE_CXX_STANDARD_REQUIRED ON)
set(CMAKE_CXX_EXTENSIONS OFF)

set(_common_compile_options -Wno-deprecated-declarations)

set(EXECUTORCH_BUILD_EXECUTOR_RUNNER ON)
set(EXECUTORCH_BUILD_EXTENSION_DATA_LOADER ON)
set(EXECUTORCH_BUILD_EXTENSION_MODULE ON)
set(EXECUTORCH_BUILD_EXTENSION_RUNNER_UTIL ON)
set(EXECUTORCH_BUILD_EXTENSION_TENSOR ON)
set(EXECUTORCH_BUILD_XNNPACK ON)

add_subdirectory("executorch")

add_executable(exec main.cpp)

target_link_libraries(
  exec
  PRIVATE executorch
          extension_module_static
          extension_tensor
          portable_kernels
          portable_ops_lib
          executorch_core
          #optimized_native_cpu_ops_lib
          xnnpack_backend)
target_link_options(exec PRIVATE -Wl,--whole-archive,--allow-multiple-definition)

It works for me, but in my project I need to build executorch outside my project

cc @larryliu0820 @manuelcandales

[larryliu0820]

Or actually - @GregoryComer can you comment on this one?

[larryliu0820]

target_link_libraries(
    exec PRIVATE
  ${EXECUTORCH_LIBRARIES}
)

Instead of linking all the libraries, can you link individual libraries?

[ismukhin]

EXECUTORCH_LIBRARIES has this variables in my build:

executorch
executorch_core
portable_kernels
extension_data_loader
portable_ops_lib
extension_module
extension_module_static
extension_tensor
extension_threadpool
xnnpack_backend
XNNPACK
microkernels-prod
cpuinfo
pthreadpool
optimized_kernels
cpublas
eigen_blas
optimized_ops_lib
optimized_native_cpu_ops_lib

I did this in cmake and this works for me:

list(REMOVE_ITEM EXECUTORCH_LIBRARIES extension_module)
list(REMOVE_ITEM EXECUTORCH_LIBRARIES optimized_kernels)
list(REMOVE_ITEM EXECUTORCH_LIBRARIES optimized_ops_lib)
list(REMOVE_ITEM EXECUTORCH_LIBRARIES optimized_native_cpu_ops_lib)

But as I understand it, the library extension_module is the same thing as extension_module_static, only dynamic and when linking it without static lib, the following error appears:

E 00:00:00.000204 executorch:operator_registry.cpp:85] Re-registering aten::sym_size.int, from NOT_SUPPORTED
E 00:00:00.000224 executorch:operator_registry.cpp:86] key: (null), is_fallback: true
F 00:00:00.000225 executorch:operator_registry.cpp:106] In function register_kernels(), assert failed (false): Kernel registration failed with error 18, see error log for details.
Aborted (core dumped)

How can this be explained?