SafeIntOnOverflow() Integer overflow error when inferencing on too many samples with Python #18905
Labels
core runtime
issues related to core runtime
Comments
|
Are you able to share a model that this occurs with? |
skottmckay
added a commit
that referenced
this issue
Dec 27, 2023
…y return a nullptr. This is inconsistent as an actual memory allocation failure throws. An overflow would typically be due to bad input so an exception makes more sense given that. Change to throw so code using MakeUniquePtr* and AllocArray* doesn't need to check for nullptr. Add some extra info to the log message to help debugging. Should help with #18905 by avoiding the invalid attempted usage of a nullptr from the allocation. Extra info _might_ help with figuring out where the overflow is coming from which is the real issue.
skottmckay
added a commit
that referenced
this issue
Jan 2, 2024
…#18941) ### Description <!-- Describe your changes. --> If we fail to calculate the buffer size (due to overflow) we currently return a nullptr. This is inconsistent as an actual memory allocation failure throws. An overflow would typically be due to bad input so an exception makes more sense given that. Change to throw so code using MakeUniquePtr* and AllocArray* doesn't need to check for nullptr. Add some extra info to the log message to help debugging. ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. --> Should help with #18905 by avoiding the invalid attempted usage of a nullptr from the allocation. Extra info _might_ help with figuring out where the overflow is coming from which is the real issue.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Describe the issue
When inferencing with too many samples in Python, I get this error on Ubuntu:
[E:onnxruntime:Default, allocator.cc:36 operator()] /onnxruntime_src/onnxruntime/core/common/safeint.h:17 static void SafeIntExceptionHandler<onnxruntime::OnnxRuntimeException>::SafeIntOnOverflow() Integer overflowWhen running the same script on an M1 Mac, it appears as a
SIGABRTwithlibc++abi terminatingmessage.I'm able to mitigate this by inferencing in batches and concatenating the results.
To reproduce
Inference with a large number of samples (low millions?). Maybe > 32767 is sufficient, since that's the max for int16, but I haven't verified that.
Urgency
No response
Platform
Linux
OS Version
Ubuntu
ONNX Runtime Installation
Released Package
ONNX Runtime Version or Commit ID
1.16.3
ONNX Runtime API
Python
Architecture
X64
Execution Provider
Default CPU
Execution Provider Library Version
No response
The text was updated successfully, but these errors were encountered: