In the Linux kernel, the following vulnerability has been resolved: bpf: tcp: Fix use-after-free in bpf_iter_tcp_established_batch()… (CVE-2026-74714)
A use-after-free vulnerability in the Linux kernel's BPF TCP iterator (bpf_iter_tcp_established_batch) was resolved. The flaw arises from incorrect reference counting of TCP request sockets during iteration, leading to a potential use-after-free condition. This occurs when a socket's reference count is overwritten improperly, causing the socket to be freed while still accessible. The issue can be reproduced with local TCP connections under specific conditions. The fix involves using safer reference counting methods and adjusting iteration logic to avoid accessing freed sockets.
AI Analysis
Technical Summary
CVE-2026-74714 is a use-after-free vulnerability in the Linux kernel's BPF TCP iterator function bpf_iter_tcp_established_batch(). The root cause is that the function uses sock_hold() to increment the socket reference count while holding a bucket lock, assuming the lock guarantees a positive reference count. However, for TCP request sockets, the reference count can be zero at that time, leading to an addition on zero and a use-after-free when the reference count is later overwritten. This causes the socket to be freed while still reachable, potentially leading to memory corruption. The vulnerability is reproducible with tcp_syncookies disabled and concurrent connect/close operations. The patch replaces sock_hold() with refcount_inc_not_zero() and modifies iteration logic to skip sockets that fail the increment, preventing use-after-free.
Potential Impact
The vulnerability allows a use-after-free condition in the kernel, which can lead to memory corruption. While the exact impact is not detailed in the provided data, use-after-free bugs in kernel code can potentially be exploited to cause system crashes or privilege escalation. No known exploits in the wild have been reported. The issue affects kernel stability and security under certain TCP connection workloads.
Mitigation Recommendations
A fix is available and has been applied to the Linux kernel to address this vulnerability. The patch uses refcount_inc_not_zero() instead of sock_hold() to safely increment reference counts and adjusts iteration logic to avoid accessing freed sockets. Users and administrators should update to the fixed kernel version as provided by their Linux distribution vendors. Patch status is not explicitly stated in the input, so check the vendor advisory for current remediation guidance.
In the Linux kernel, the following vulnerability has been resolved: bpf: tcp: Fix use-after-free in bpf_iter_tcp_established_batch()… (CVE-2026-74714)
Description
A use-after-free vulnerability in the Linux kernel's BPF TCP iterator (bpf_iter_tcp_established_batch) was resolved. The flaw arises from incorrect reference counting of TCP request sockets during iteration, leading to a potential use-after-free condition. This occurs when a socket's reference count is overwritten improperly, causing the socket to be freed while still accessible. The issue can be reproduced with local TCP connections under specific conditions. The fix involves using safer reference counting methods and adjusting iteration logic to avoid accessing freed sockets.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
CVE-2026-74714 is a use-after-free vulnerability in the Linux kernel's BPF TCP iterator function bpf_iter_tcp_established_batch(). The root cause is that the function uses sock_hold() to increment the socket reference count while holding a bucket lock, assuming the lock guarantees a positive reference count. However, for TCP request sockets, the reference count can be zero at that time, leading to an addition on zero and a use-after-free when the reference count is later overwritten. This causes the socket to be freed while still reachable, potentially leading to memory corruption. The vulnerability is reproducible with tcp_syncookies disabled and concurrent connect/close operations. The patch replaces sock_hold() with refcount_inc_not_zero() and modifies iteration logic to skip sockets that fail the increment, preventing use-after-free.
Potential Impact
The vulnerability allows a use-after-free condition in the kernel, which can lead to memory corruption. While the exact impact is not detailed in the provided data, use-after-free bugs in kernel code can potentially be exploited to cause system crashes or privilege escalation. No known exploits in the wild have been reported. The issue affects kernel stability and security under certain TCP connection workloads.
Mitigation Recommendations
A fix is available and has been applied to the Linux kernel to address this vulnerability. The patch uses refcount_inc_not_zero() instead of sock_hold() to safely increment reference counts and adjusts iteration logic to avoid accessing freed sockets. Users and administrators should update to the fixed kernel version as provided by their Linux distribution vendors. Patch status is not explicitly stated in the input, so check the vendor advisory for current remediation guidance.
Technical Details
- Gcve Source
- db.gcve.eu
- Osv Id
- GHSA-cqmm-fv72-p7f7
- Osv Schema Version
- 1.4.0
- Aliases
- ["CVE-2026-74714"]
- Ecosystems
- []
- Database Specific Severity
- null
- Cvss Version
- null
Threat ID: 6a8a27f0acd9273b499bc6ee
Added to database: 08/22/2026, 22:51:28 UTC
Last enriched: 08/23/2026, 00:54:57 UTC
Last updated: 08/23/2026, 02:32:09 UTC
Views: 2
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.