CWE-120

NVIDIA BMC contains a vulnerability in IPMI handler, where an authorized attacker can cause a buffer overflow and cause a denial of service or gain code execution.

NVIDIA vGPU software contains a vulnerability in the Virtual GPU Manager (vGPU plugin), where an input index is not validated, which may lead to buffer overrun, which in turn may cause data tampering, information disclosure, or denial of service.

NVIDIA vGPU software contains a vulnerability in the Virtual GPU Manager (vGPU plugin), where an input index is not validated, which may lead to buffer overrun, which in turn may cause data tampering, information disclosure, or denial of service.

XStream serializes Java objects to XML and back again. Versions prior to 1.4.20 may allow a remote attacker to terminate the application with a stack overflow error, resulting in a denial of service only via manipulation the processed input stream. The attack uses the hash code implementation for collections and maps to force recursive hash calculation causing a stack overflow. This issue is patched in version 1.4.20 which handles the stack overflow and raises an InputManipulationException instead. A potential workaround for users who only use HashMap or HashSet and whose XML refers these only as default map or set, is to change the default implementation of java.util.Map and java.util per the code example in the referenced advisory. However, this implies that your application does not care about the implementation of the map and all elements are comparable.

TensorFlow is an open source platform for machine learning. The reference kernel of the `CONV_3D_TRANSPOSE` TensorFlow Lite operator wrongly increments the data_ptr when adding the bias to the result. Instead of `data_ptr += num_channels;` it should be `data_ptr += output_num_channels;` as if the number of input channels is different than the number of output channels, the wrong result will be returned and a buffer overflow will occur if num_channels > output_num_channels. An attacker can craft a model with a specific number of input channels. It is then possible to write specific values through the bias of the layer outside the bounds of the buffer. This attack only works if the reference kernel resolver is used in the interpreter. We have patched the issue in GitHub commit 72c0bdcb25305b0b36842d746cc61d72658d2941. The fix will be included in TensorFlow 2.11. We will also cherrypick this commit on TensorFlow 2.10.1, 2.9.3, and TensorFlow 2.8.4, as these are also affected and still in supported range.

An integer overflow and buffer overflow issues were found in the ACPI Error Record Serialization Table (ERST) device of QEMU in the read_erst_record() and write_erst_record() functions. Both issues may allow the guest to overrun the host buffer allocated for the ERST memory device. A malicious guest could use these flaws to crash the QEMU process on the host.