549 matches found
CVE-2023-53438 x86/MCE: Always save CS register on AMD Zen IF Poison errors
In the Linux kernel, the following vulnerability has been resolved: x86/MCE: Always save CS register on AMD Zen IF Poison errors The Instruction Fetch IF units on current AMD Zen-based systems do not guarantee a synchronous MC is delivered for poison consumption errors. Therefore,...
CVE-2023-53438
CVE-2023-53438 : In the Linux kernel, the AMD Zen IF poison error path could misclassify context due to non-synchronous delivery of poison by the IF unit. The issue is mitigated by adding a quirk to always save the Code Segment (CS) register when poison is consumed from the IF unit banks, ensurin...
CVE-2023-53438 x86/MCE: Always save CS register on AMD Zen IF Poison errors
In the Linux kernel, the following vulnerability has been resolved: x86/MCE: Always save CS register on AMD Zen IF Poison errors The Instruction Fetch IF units on current AMD Zen-based systems do not guarantee a synchronous MC is delivered for poison consumption errors. Therefore,...
PT-2025-38457
Name of the Vulnerable Software and Affected Versions Linux kernel affected versions not specified Description The Linux kernel contains a flaw within the x86/MCE subsystem related to handling Machine Check Exceptions MCE on AMD Zen-based systems. Specifically, the Instruction Fetch IF units may...
mm/vmscan: don't try to reclaim hwpoison folio
...
Linux Distros Unpatched Vulnerability : CVE-2022-32091
The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - MariaDB v10.7 was discovered to contain an use-after-poison in in interceptormemset at /libsanitizer/sanitizercommon/sanitizercommoninterceptors.inc...
Linux Distros Unpatched Vulnerability : CVE-2022-32081
The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - MariaDB v10.4 to v10.7 was discovered to contain an use-after-poison in prepareinplaceaddvirtual at /storage/innobase/handler/handler0alter.cc. CVE-2022-32081...
Non-Omniscient Backdoor Injection with a Single Poison Sample: Proving the One-Poison Hypothesis for Linear Regression and Linear Classification
Backdoor injection attacks are a threat to machine learning models that are trained on large data collected from untrusted sources; these attacks enable attackers to inject malicious behavior into the model that can be triggered by specially crafted inputs. Prior work has established bounds on th...
Linux Distros Unpatched Vulnerability : CVE-2025-5994
The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - A multi-vendor cache poisoning vulnerability named 'Rebirthday Attack' has been discovered in caching resolvers that support EDNS Client Subnet ECS. Unbound is...
A Bayesian Incentive Mechanism for Poison-Resilient Federated Learning
Federated learning FL enables collaborative model training across decentralized clients while preserving data privacy. However, its open-participation nature exposes it to data-poisoning attacks, in which malicious actors submit corrupted model updates to degrade the global model. Existing defens...
AZL-72587 CVE-2025-38334 affecting package kernel 5.15.200.1-1
In the Linux kernel, the following vulnerability has been resolved: x86/sgx: Prevent attempts to reclaim poisoned pages TL;DR: SGX page reclaim touches the page to copy its contents to secondary storage. SGX instructions do not gracefully handle machine checks. Despite this, the existing SGX code...
Juju allows arbitrary executable uploads via authenticated endpoint without authorization
Summary You can affect the agent binaries used in a Juju controller and the code that is run in the binaries by simply having a user account on a controller. You aren't required to have a model or any permissions. This just requires a user account in the controller database. Details Because of th...
PT-2025-33780
Name of the Vulnerable Software and Affected Versions: Linux kernel affected versions not specified Description: The Linux kernel contains a flaw in the RDMA/hns subsystem where rsv qp may be double destroyed during an error condition. This occurs first in free mr init and then in hns roce exit...
OSV-2025-510 Use-after-poison in llvm::BitstreamCursor::readRecord
OSS-Fuzz report: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=428525443 Crash type: Use-after-poison READ 1 Crash state: llvm::BitstreamCursor::readRecord BitcodeReader::parseModule llvm::BitcodeModule::getModuleImpl...
PDLRecover: Privacy-preserving Decentralized Model Recovery with Machine Unlearning
Decentralized learning is vulnerable to poison attacks, where malicious clients manipulate local updates to degrade global model performance. Existing defenses mainly detect and filter malicious models, aiming to prevent a limited number of attackers from corrupting the global model. However,...
Astra Linux – Vulnerability in Linux 6.12
In the Linux kernel, the following vulnerabilities have been resolved: x86/mce: Use iscopyfromuser to determine the copyfromuser context Patches in the series “mm/hwpoison: Fix regressions in memory failure handling”, version 4. 1. What is being attempted? This patchset addresses two critical...
BIT-MARIADB-MIN-2022-32091
MariaDB v10.7 was discovered to contain an use-after-poison in in interceptormemset at /libsanitizer/sanitizercommon/sanitizercommoninterceptors.inc...
BIT-MARIADB-MIN-2022-32081
MariaDB v10.4 to v10.7 was discovered to contain an use-after-poison in prepareinplaceaddvirtual at /storage/innobase/handler/handler0alter.cc...
Poison in the Well: Feature Embedding Disruption in Backdoor Attacks
Backdoor attacks embed malicious triggers into training data, enabling attackers to manipulate neural network behavior during inference while maintaining high accuracy on benign inputs. However, existing backdoor attacks face limitations manifesting in excessive reliance on training data, poor...
