skottmckay
commented
1 week ago Hard to say without a stack trace. with symbol names
ORT will do most allocations during model initialization and the first inference. After that it's using a cache for memory so segfaults would typically be an out-of-memory scenario or bad input (e.g. input tensor is freed while ORT is using it).
If you're building from source can you build a debug version? May need to ensure the Android build doesn't strip the binary of symbols though as typically it.
Does the issue happen if you run on the Android emulator? Would be easier to debug if it did.
Another option would be to copy onnxruntime_perf_test using adb to the phone (use /data/local/tmp), along with the model, and run. you can specify the number of iterations or amount of time to run for, and it can generate dummy input data.
laurenspriem
Windsander
Describe the issue
I am getting a segmentation fault (SIGSEGV) after repeated inference runs on mobile, crashing the app. The issue only comes up after running inference for more than 300 times, but it consistently comes up after that. For context, I am using ORT in a Flutter app through FFI.
Error logs
To reproduce
The issue is reproducible by letting the app continuously run inference. Since this is happening in the app, it's a bit hard to give a clear and easy MRE. If nothing comes up from the error logs alone I'll try to create a dummy app that reproduces the issue and share the code for it here.
Urgency
I don't know how urgent this issue is to ORT, but for our app it's quite urgent.
Platform
Android
OS Version
Android 14 (and other versions)
ONNX Runtime Installation
Built from Source
Compiler Version (if 'Built from Source')
No response
Package Name (if 'Released Package')
None
ONNX Runtime Version or Commit ID
v1.15.0
ONNX Runtime API
C++/C
Architecture
ARM64
Execution Provider
Default CPU
Execution Provider Library Version
No response