Loading vulnerability…
CVE-2024-56675 is a use-after-free vulnerability in the Linux kernel BPF tracing subsystem. Uprobes execute attached programs through bpf_prog_run_array_uprobe() while protected by tasks-trace-RCU. However, non-sleepable BPF programs may be attached to uprobes and released through normal RCU by __bpf_prog_put_noref(). Completion of a normal RCU grace period does not guarantee completion of a tasks-trace-RCU grace period, allowing an uprobe execution path to access a freed bpf_prog. The upstream fix explicitly waits for a tasks-trace-RCU grace period after removing a BPF-program attachment from a perf event.
Mallory correlates every CVE against your assets, your vendors, and active adversary campaigns. Know which vulnerabilities matter for you, not just which ones are loud.
What it means. What to do now. Patch path, mitigations, and the assume-compromise checklist.
What an attacker gets, and what they’ve been doing with it.
If you can’t patch tonight, do this now.
Patch, then assume compromise.
No public exploits tracked yet. Mallory keeps watching.
No public exploit code observed for this vulnerability.
Products and vendors Mallory has correlated with this vulnerability. Open in Mallory to drill down to specific CPE configurations and version ranges.
Vendor-confirmed product mapping. Mallory continuously reconciles this list against your asset inventory.
5 sources tracked across advisories, community write-ups, and news. New activity surfaces here as Mallory finds it.
Linux kernel BPF use-after-free vulnerability.
Linux kernel BPF use-after-free vulnerability involving mismatched RCU handling.
A Linux kernel eBPF use-after-free vulnerability caused by a mismatch between normal RCU cleanup of non-sleepable BPF programs and tasks-trace-RCU protection used for uprobe execution. A BPF program attached to a perf event could be freed before the relevant tasks-trace-RCU grace period completed.
A Linux kernel eBPF use-after-free vulnerability caused by mismatched RCU protection for non-sleepable BPF programs attached to uprobes. The flaw may affect confidentiality, integrity, and availability. Red Hat rates it Moderate (CVSS 6.7) because its default RHEL kernel configuration disables unprivileged eBPF, requiring CAP_SYS_ADMIN or root to exploit.
Query your assets running an affected version, and investigate the blast radius.
Every observed campaign linking this CVE to a named adversary.
Malware families riding this exploit, with evidence and IOCs.
YARA, Sigma, Snort, and vendor rules, auto-deployed to your SIEM.
Cross-references every affected SKU, including bundled OEM variants.
Community discussion across Reddit, Mastodon, and other social sources.