Missing operator: [3] cadence::quantized_relu.per_tensor_out when following build-run-xtensa tutorial

[ELR11C]

🐛 Describe the bug

I have installed executorch using WSL (Ubuntu 22.04) and am running into issues when following the tutorial found here. Specifically, I am running python3 -m examples.cadence.models.rnnt_predictor in the executorch folder using my venv, and I obtain the following error log after the CXX executable cadence_runner is built:

[100%] Linking CXX executable cadence_runner
[100%] Built target cadence_runner
Built cmake-out/backends/cadence/cadence_runner
[INFO 2025-03-03 17:19:19,965 executor.py:129] ./cmake-out/backends/cadence/cadence_runner --bundled_program_path=/tmp/tmp5meia7b1/CadenceDemoModel.bpte --etdump_path=/tmp/tmp5meia7b1/etdump.etdp --debug_output_path=/tmp/tmp5meia7b1/debug_output.bin --dump_outputs=true
I 00:00:00.001455 executorch:example_runner.cpp:145] Model file /tmp/tmp5meia7b1/CadenceDemoModel.bpte is loaded.
I 00:00:00.001471 executorch:example_runner.cpp:154] Running method forward
I 00:00:00.001473 executorch:example_runner.cpp:201] Setting up planned buffer 0, size 20496.
E 00:00:00.001498 executorch:operator_registry.cpp:252] kernel 'cadence::quantized_relu.per_tensor_out' not found.
E 00:00:00.001500 executorch:operator_registry.cpp:253] dtype: 1 | dim order: [
E 00:00:00.001501 executorch:operator_registry.cpp:253] 0,
E 00:00:00.001503 executorch:operator_registry.cpp:253] 1,
E 00:00:00.001504 executorch:operator_registry.cpp:253] 2,
E 00:00:00.001505 executorch:operator_registry.cpp:253] ]
E 00:00:00.001506 executorch:operator_registry.cpp:253] dtype: 0 | dim order: [
E 00:00:00.001507 executorch:operator_registry.cpp:253] 0,
E 00:00:00.001508 executorch:operator_registry.cpp:253] 1,
E 00:00:00.001510 executorch:operator_registry.cpp:253] 2,
E 00:00:00.001511 executorch:operator_registry.cpp:253] ]
E 00:00:00.001512 executorch:operator_registry.cpp:253] dtype: 0 | dim order: [
E 00:00:00.001513 executorch:operator_registry.cpp:253] 0,
E 00:00:00.001514 executorch:operator_registry.cpp:253] 1,
E 00:00:00.001515 executorch:operator_registry.cpp:253] 2,
E 00:00:00.001516 executorch:operator_registry.cpp:253] ]
E 00:00:00.001518 executorch:method.cpp:724] Missing operator: [3] cadence::quantized_relu.per_tensor_out
E 00:00:00.001521 executorch:method.cpp:944] There are 1 instructions don't have corresponding operator registered. See logs for details
F 00:00:00.001527 executorch:example_runner.cpp:220] In function main(), assert failed (method.ok()): Loading of method forward failed with status 0x14
Traceback (most recent call last):
File "/home/elriic/executorch_venv/lib/python3.10/site-packages/executorch/backends/cadence/runtime/executor.py", line 94, in execute
return _execute_subprocess(args)
File "/home/elriic/executorch_venv/lib/python3.10/site-packages/executorch/backends/cadence/runtime/executor.py", line 78, in _execute_subprocess
raise subprocess.CalledProcessError(p.returncode, p.args, stdout, stderr)
subprocess.CalledProcessError: Command '['./cmake-out/backends/cadence/cadence_runner', '--bundled_program_path=/tmp/tmp5meia7b1/CadenceDemoModel.bpte', '--etdump_path=/tmp/tmp5meia7b1/etdump.etdp', '--debug_output_path=/tmp/tmp5meia7b1/debug_output.bin', '--dump_outputs=true']' died with <Signals.SIGABRT: 6>.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/usr/lib/python3.10/runpy.py", line 196, in _run_module_as_main
return _run_code(code, main_globals, None,
File "/usr/lib/python3.10/runpy.py", line 86, in _run_code
exec(code, run_globals)
File "/home/elriic/executorch/examples/cadence/models/rnnt_predictor.py", line 69, in
export_model(model, example_inputs)
File "/home/elriic/executorch_venv/lib/python3.10/site-packages/executorch/backends/cadence/aot/export_example.py", line 117, in export_model
runtime.run_and_compare(
File "/home/elriic/executorch_venv/lib/python3.10/site-packages/executorch/backends/cadence/runtime/runtime.py", line 220, in run_and_compare
outputs = run(executorch_prog, inputs, ref_outputs, working_dir)
File "/home/elriic/executorch_venv/lib/python3.10/site-packages/executorch/backends/cadence/runtime/runtime.py", line 158, in run
executor()
File "/home/elriic/executorch_venv/lib/python3.10/site-packages/executorch/backends/cadence/runtime/executor.py", line 130, in call
execute(args)
File "/home/elriic/executorch_venv/lib/python3.10/site-packages/executorch/backends/cadence/runtime/executor.py", line 97, in execute
raise RuntimeError(
RuntimeError: Failed to execute. Use the following to debug:
fdb ./cmake-out/backends/cadence/cadence_runner --bundled_program_path=/tmp/tmp5meia7b1/CadenceDemoModel.bpte --etdump_path=/tmp/tmp5meia7b1/etdump.etdp --debug_output_path=/tmp/tmp5meia7b1/debug_output.bin --dump_outputs=true

Versions

PyTorch version: 2.7.0.dev20250131+cpu
Is debug build: False
CUDA used to build PyTorch: None
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.5 LTS (x86_64)
GCC version: (Ubuntu 12.3.0-1ubuntu1~22.04) 12.3.0
Clang version: Could not collect
CMake version: version 3.31.4
Libc version: glibc-2.35

Python version: 3.10.12 (main, Feb 4 2025, 14:57:36) [GCC 11.4.0] (64-bit runtime)
Python platform: Linux-5.15.167.4-microsoft-standard-WSL2-x86_64-with-glibc2.35
Is CUDA available: False
CUDA runtime version: No CUDA
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA
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): 8
On-line CPU(s) list: 0-7
Vendor ID: GenuineIntel
Model name: Intel(R) Core(TM) i5-8350U CPU @ 1.70GHz
CPU family: 6
Model: 142
Thread(s) per core: 2
Core(s) per socket: 4
Socket(s): 1
Stepping: 10
BogoMIPS: 3791.99
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 arch_perfmon rep_good nopl xtopology cpuid pni pclmulqdq vmx ssse3 fma cx16 pdcm pcid sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi ept vpid ept_ad fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap clflushopt xsaveopt xsavec xgetbv1 xsaves md_clear flush_l1d arch_capabilities
Virtualization: VT-x
Hypervisor vendor: Microsoft
Virtualization type: full
L1d cache: 128 KiB (4 instances)
L1i cache: 128 KiB (4 instances)
L2 cache: 1 MiB (4 instances)
L3 cache: 6 MiB (1 instance)
Vulnerability Gather data sampling: Unknown: Dependent on hypervisor status
Vulnerability Itlb multihit: KVM: Mitigation: VMX disabled
Vulnerability L1tf: Mitigation; PTE Inversion; VMX conditional cache flushes, SMT vulnerable
Vulnerability Mds: Mitigation; Clear CPU buffers; SMT Host state unknown
Vulnerability Meltdown: Mitigation; PTI
Vulnerability Mmio stale data: Mitigation; Clear CPU buffers; SMT Host state unknown
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Mitigation; 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; IBRS; IBPB conditional; STIBP conditional; RSB filling; PBRSB-eIBRS Not affected; BHI SW loop, KVM SW loop
Vulnerability Srbds: Unknown: Dependent on hypervisor status
Vulnerability Tsx async abort: Mitigation; Clear CPU buffers; SMT Host state unknown

Versions of relevant libraries:
[pip3] executorch==0.5.0a0+1bc0699
[pip3] numpy==2.0.0
[pip3] torch==2.7.0.dev20250131+cpu
[pip3] torchao==0.10.0+git7d879462
[pip3] torchaudio==2.6.0.dev20250131+cpu
[pip3] torchsr==1.0.4
[pip3] torchvision==0.22.0.dev20250131+cpu
[conda] Could not collect

cc @mcremon-meta

[iseeyuan]

Since Matthias is on leave, @zonglinpeng and @tarun292 could you help on this issue?

[zonglinpeng]

codegen yaml may be misaligned due to recent change. Let me check and will update

[zonglinpeng]

quantized_relu_per_tensor_out is missing in cpu flow. Fixed in #8907
@ELR11C Thanks for capturing this