ARM backend: --model_input flag in run.sh is not passed to the aot_arm_compiler

[Juanfi8]

🐛 Describe the bug

The --model_input flag in the run.sh is not passed to the arm compiler.

Versions

Collecting environment information...
PyTorch version: N/A
Is debug build: N/A
CUDA used to build PyTorch: N/A
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.5 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 3.31.6
Libc version: glibc-2.35

Python version: 3.10.0 (default, Mar 10 2025, 16:20:48) [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: N/A
CUDA runtime version: 11.5.119
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: GPU 0: Quadro T2000
Nvidia driver version: 538.78
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.8.2.4
/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.2.4
/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.2.4
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.2.4
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.2.4
/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.2.4
/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.2.4
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: N/A

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): 12
On-line CPU(s) list: 0-11
Vendor ID: GenuineIntel
Model name: Intel(R) Core(TM) i7-10850H CPU @ 2.70GHz
CPU family: 6
Model: 165
Thread(s) per core: 2
Core(s) per socket: 6
Socket(s): 1
Stepping: 2
BogoMIPS: 5424.01
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 x2apic movbe popcnt 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 bmi1 avx2 smep bmi2 erms invpcid 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: 192 KiB (6 instances)
L1i cache: 192 KiB (6 instances)
L2 cache: 1.5 MiB (6 instances)
L3 cache: 12 MiB (1 instance)
Vulnerability Gather data sampling: Unknown: Dependent on hypervisor status
Vulnerability Itlb multihit: KVM: Mitigation: VMX disabled
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Mitigation; Clear CPU buffers; SMT Host state unknown
Vulnerability Reg file data sampling: Not affected
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 SW loop, KVM SW loop
Vulnerability Srbds: Unknown: Dependent on hypervisor status
Vulnerability Tsx async abort: Not affected

Versions of relevant libraries:
[pip3] nvidia-cublas-cu12==12.4.5.8
[pip3] nvidia-cuda-cupti-cu12==12.4.127
[pip3] nvidia-cuda-nvrtc-cu12==12.4.127
[pip3] nvidia-cuda-runtime-cu12==12.4.127
[pip3] nvidia-cudnn-cu12==9.1.0.70
[pip3] nvidia-cufft-cu12==11.2.1.3
[pip3] nvidia-curand-cu12==10.3.5.147
[pip3] nvidia-cusolver-cu12==11.6.1.9
[pip3] nvidia-cusparse-cu12==12.3.1.170
[pip3] nvidia-cusparselt-cu12==0.6.2
[pip3] nvidia-nccl-cu12==2.21.5
[pip3] nvidia-nvjitlink-cu12==12.4.127
[pip3] nvidia-nvtx-cu12==12.4.127
[pip3] onnx==1.17.0
[pip3] onnxruntime==1.20.1
[pip3] onnxsim==0.4.36
[pip3] torchsr==1.0.4
[pip3] triton==3.2.0
[conda] Could not collect

cc @digantdesai @freddan80 @per @zingo @oscarandersson8218

[gggekov]

Hi @Juanfi8,
Thanks for raising the issue. What is your exact command with run.sh and how is the problem manifested in your environment? The model_input is passed here(https://github.com/pytorch/executorch/blob/main/examples/arm/aot_arm_compiler.py#L70) and we use that model input when running with bundleIO, it would be useful if you could provide more information to understand the problem you face.

[Juanfi8]

Hi, sorry if I was unclear in my first comment. The run.sh script in the examples/arm folder allows the user to specify a different model_input than the one defined in the model file. As you mentioned, the expected behavior of the script is to pass the model_input as an argument to the aot_arm_compiler, but it doesn't seem to be implemented.

executorch/examples/arm/run.sh

Line 204 in 753da9a

ARM_AOT_CMD="python3 -m examples.arm.aot_arm_compiler --model_name=${model} --target=${target} ${model_compiler_flags} --intermediate=${output_folder} --output=${pte_file} --so_library=$SO_LIB --system_config=${system_config} --memory_mode=${memory_mode} $bundleio_flag"

[zingo]

Thanks for clarifying, we will look into it when getting back at work next week. Good catch.

If you want to add a fix/PR and get the fame and glory for your find that is also welcome.

[zingo]

Also as a note, when I added this I added support for .pt files and if you think other files formats would be useful you can just follow the flag where it is used in examples/arm/aot_arm_compiler.py (around line 70) and it's probably easy for you to add support for more formats if it makes sense. :)

[zingo]

Thanks for the PR, you where faster then our internal review process on my patch, and your patch is a bit better with the updated help text :)