medium

CVE-2026-44512

PyPI · onnx

Summary

ONNX has Null Pointer Dereference in Upsample Version Converter Adapter (Zero Inputs)

Severity
medium
CVSS
5.5
EPSS
0.2% (p9)
CWE
CWE-476
Also known as
GHSA-hwpq-hmq9-wj77
Published
2026-07-07
Updated
2026-07-07

Advisory details

Summary

Null pointer dereference (SIGSEGV) in Upsample_6_7::adapt_upsample_6_7() (onnx/version_converter/adapters/upsample_6_7.h:31) when convert_version() processes a model with an Upsample node that has zero inputs. The adapter accesses node->inputs()[0]->sizes() without checking input count. 107-byte PoC crashes on Release build.

This is the same class of bug as the Cast adapter advisory (separate report) but in a different adapter, different file, and different operator.

Details

The Upsample 6→7 adapter validates attributes but not inputs:

// upsample_6_7.h:20-33
void adapt_upsample_6_7(..., Node* node) const {
    ONNX_ASSERTM(
        node->hasAttribute(width_scale_symbol) && node->hasAttribute(height_scale_symbol),
        "...")  // Attribute check PASSES

    auto width_scale = node->f(width_scale_symbol);
    auto height_scale = node->f(height_scale_symbol);

    auto input_shape = node->inputs()[0]->sizes();
    //                 ^^^^^^^^^^^^^^^^^^^^
    //                 OOB when inputs().size() == 0 → SIGSEGV
}

The PoC has an Upsample node at opset 6 with the required width_scale and height_scale attributes but zero inputs. The attribute assertions pass, then node->inputs()[0] on an empty ArrayRef:

An Upsample node with zero inputs passes graphProtoToGraph() because the import code only resolves input names present in the protobuf.

PoC

import base64
import onnx
from onnx import version_converter

poc_b64 = "CAI6YQo8EgFZIghVcHNhbXBsZSoVCgt3aWR0aF9zY2FsZRUAAABAoAEBKhYKDGhlaWdodF9zY2FsZRUAAABAoAEBEgR0ZXN0YhsKAVkSFgoUCAESEAoCCAEKAggBCgIIBAoCCARCBAoAEAY="

model = onnx.load_from_string(base64.b64decode(poc_b64))

# CRASHES — Upsample_6_7 adapter dereferences empty inputs array
version_converter.convert_version(model, 7)  # SIGSEGV

107-byte PoC. Confirmed SIGSEGV on both onnx 1.21.0 (pip) and 1.22.0 (source build).

Impact

Any application that uses onnx.version_converter.convert_version() on untrusted models is vulnerable. This includes model conversion pipelines and tools that auto-upgrade opset versions for compatibility. The crash is unrecoverable (SIGSEGV).

This vulnerability is part of a systemic pattern across multiple version converter adapters. A full audit of all ~45 adapters was performed as part of the fix; eight adapters were found with the same class of unguarded indexed access (cast_9_8, softmax_12_13, softmax_13_12, upsample_6_7, upsample_9_10, group_normalization_20_21, broadcast_forward_compatibility, upsample_9_8) and all have been fixed in PR #7813.

References

Related advisories

Is your project exposed to this? Stateward checks every dependency on every pull request and flags it only if your code actually reaches it.

Check my repo

Summarize with AI

ChatGPTClaudePerplexity

Sources: CISA KEV (public domain), OSV.dev & GitHub Advisory Database (CC-BY-4.0), FIRST EPSS, NVD/CWE (public domain). Served live from the Stateward advisory database.