tensorflow
Affected
- introduced
0, fixed2.5.3(ecosystem)
Fixed versions
2.5.3
Fix
- Fixed version
2.5.3pip install --upgrade tensorflow==2.5.3
Advisory
Also known as: BIT-tensorflow-2022-21725, CVE-2022-21725, PYSEC-2022-104, PYSEC-2022-49, PYSEC-2026-3242
Division by zero in Tensorflow
Impact
The estimator for the cost of some convolution operations can be made to execute a division by 0:
import tensorflow as tf
@tf.function
def test():
y=tf.raw_ops.AvgPoolGrad(
orig_input_shape=[1,1,1,1],
grad=[[[[1.0],[1.0],[1.0]]],[[[2.0],[2.0],[2.0]]],[[[3.0],[3.0],[3.0]]]],
ksize=[1,1,1,1],
strides=[1,1,1,0],
padding='VALID',
data_format='NCHW')
return y
test()
The function fails to check that the stride argument is stricly positive:
int64_t GetOutputSize(const int64_t input, const int64_t filter,
const int64_t stride, const Padding& padding) {
// Logic for calculating output shape is from GetWindowedOutputSizeVerbose()
// function in third_party/tensorflow/core/framework/common_shape_fns.cc.
if (padding == Padding::VALID) {
return (input - filter + stride) / stride;
} else { // SAME.
return (input + stride - 1) / stride;
}
}
Hence, the fix is to add a check for the stride argument to ensure it is valid.
Patches
We have patched the issue in GitHub commit 3218043d6d3a019756607643cf65574fbfef5d7a.
The fix will be included in TensorFlow 2.8.0. We will also cherrypick this commit on TensorFlow 2.7.1, TensorFlow 2.6.3, and TensorFlow 2.5.3, as these are also affected and still in supported range.
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 Yu Tian of Qihoo 360 AIVul Team.
Medium Basis: The source's own band: MODERATE
0, fixed 2.5.3 (ecosystem)2.5.3
2.5.3pip install --upgrade tensorflow==2.5.32.6.0, fixed 2.6.3 (ecosystem)2.6.3
2.6.3pip install --upgrade tensorflow==2.6.32.7.0, fixed 2.7.1 (ecosystem)2.7.1
2.7.1pip install --upgrade tensorflow==2.7.10, fixed 2.5.3 (ecosystem)2.5.3
2.5.3pip install --upgrade tensorflow-cpu==2.5.32.6.0, fixed 2.6.3 (ecosystem)2.6.3
2.6.3pip install --upgrade tensorflow-cpu==2.6.32.7.0, fixed 2.7.1 (ecosystem)2.7.1
2.7.1pip install --upgrade tensorflow-cpu==2.7.10, fixed 2.5.3 (ecosystem)2.5.3
2.5.3pip install --upgrade tensorflow-gpu==2.5.32.6.0, fixed 2.6.3 (ecosystem)2.6.3
2.6.3pip install --upgrade tensorflow-gpu==2.6.32.7.0, fixed 2.7.1 (ecosystem)2.7.1
2.7.1pip install --upgrade tensorflow-gpu==2.7.1Fix commands are the export's own generic commands. Ojuju never runs them. Check them against your setup first.
Check your own computer: guard packages check
Source: GitHub Advisory Database (reviewed advisories) · CC BY 4.0