Look up another CVE · open CVE-2026-63865 in the interactive tracker
CVE-2026-63865
bpf: Drop task_to_inode and inet_conn_established from lsm sleepable hooks
Published 2026-07-19 (3 months ago) by the Linux kernel CNA. Subsystem kernel/bpf. Debian urgency: not yet assigned.
What to do, per Debian release
Debian's own status for CVE-2026-63865 in every release this site tracks, and what each one means in practice. Every status below is the one the Debian Security Team publishes for this CVE. source
| Release | Status | Version | What this means |
|---|---|---|---|
| Debian 12bookworm | fixed | 6.1.176-1 | Update the kernel and reboot. Fixed in 6.1.176-1. A running kernel keeps the old code until the machine restarts. |
| Debian 13trixie | fixed | 6.12.94-1 | Update the kernel and reboot. Fixed in 6.12.94-1. A running kernel keeps the old code until the machine restarts. |
| Debian 14forky | fixed | 7.0.10-1 | Update the kernel and reboot. Fixed in 7.0.10-1. A running kernel keeps the old code until the machine restarts. |
| Debian unstablesid | fixed | 7.0.10-1 | Update the kernel and reboot. Fixed in 7.0.10-1. A running kernel keeps the old code until the machine restarts. |
The vulnerability
What the bug is
bpf_lsm_task_to_inode() is called under rcu_read_lock() and bpf_lsm_inet_conn_established() is called from softirq context, so neither hook can be used by sleepable LSM programs.
Where it lives
The fix touches these files. If the subsystem is one you do not use, the practical exposure is lower, though the package is still the vulnerable one.
kernel/bpf/bpf_lsm.c
- What this is
- BPF [SECURITY & LSM] (Security Audit and Enforcement using BPF)MAINTAINERS section matched on kernel/bpf/bpf_lsm.c · bpf@vger.kernel.org · Maintained
Show whether Debian builds this, per release
Not established. This site could not find the build rule for the affected file in the kernel trees Debian ships, so it has no answer about whether Debian compiles this code. Read that as unknown, never as absent.
- bookworm, trixie, forky, sid?
- ?
- not established - no Makefile entry for the affected file
Where a release above reads "not established", no build rule for the affected file was found in that release's kernel tree. That is a gap in this lookup, not a finding that the code is absent.
From the configuration Debian publishes for every kernel flavour it builds. It covers only the architectures Debian ships a linux-config package for, and nothing about a kernel you built yourself. Debian's own status above stays authoritative about the source package.MAINTAINERS
Signals
- Published
- 2026-07-19 (3 months ago)source
- CVSS
- 8.8 high
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:Hsource recompute it - Attack vector
- local (AV:L)source
- EPSS
- 0.13%, higher than 2.1% of all CVEssource
- CISA KEV
- not listedthe KEV catalogue
- Debian urgency
- not yet assignedsource
- Subsystem
- kernel/bpf
Upstream fixes
The versions the CNA records as introducing and fixing this, and the commit for each. source
| Introduced | Fixed in | Commit |
|---|---|---|
| 5.11 | 5.15.209 | 452a927cddcd |
| 5.11 | 6.1.175 | f0fc2a982817 |
| 5.11 | 6.6.141 | 26b380a3ca0b |
| 5.11 | 6.12.91 | 0d918263c9bf |
| 5.11 | 6.18.33 | 281f2a214565 |
| 5.11 | 7.0.10 | 989f1b93907d |
| 5.11 | 7.1 | beaf0e96b1da |
Why the CNA scored it this way
Quoted from the kernel CNA's own scoring file, word for word. A highlighted metric carries the most severe value CVSS v3.1 defines for it, which is what pushes the score up. source
- AV:LAttack vector, local
- Exploitation requires loading and attaching a sleepable BPF LSM program via bpf() syscalls; per kernel CNA guidance BPF subsystem issues are Local even though inet_conn_established fires during TCP receive processing.
- AC:LAttack complexity, low
- Once a sleepable BPF LSM program is attached, the attacker reliably triggers the bug by making outbound TCP connections (inet_conn_established in softirq) or accessing /proc entries (task_to_inode under rcu_read_lock) without race-dependent conditions.
- PR:LPrivileges required, low
- Loading BPF_PROG_TYPE_LSM programs requires CAP_BPF and CAP_PERFMON, both obtainable by an unprivileged user inside a user namespace (unshare -Ur) on typical Linux deployments with CONFIG_BPF_LSM enabled.
- UI:NUser interaction, none
- No victim user interaction is required; triggering occurs automatically from normal outbound TCP handshakes or /proc filesystem access after the sleepable BPF program is loaded.
- S:CScope, changed
- A container tenant with namespace CAP_BPF can load a sleepable BPF LSM on inet_conn_established or task_to_inode, and hook execution in host kernel softirq/RCU context can crash or corrupt the entire host, crossing the container-to-host security boundary.
- C:HConfidentiality impact, high
- Executing sleepable BPF helpers (e.g. bpf_copy_from_user, GFP_KERNEL allocations) inside rcu_read_lock or softirq can corrupt kernel memory before detection, and such context violations are classifiable as exploitable memory-safety failures.
- I:HIntegrity impact, high
- Sleeping while holding TCP stack or RCU-protected locks can corrupt kernel data structures and potentially be leveraged for arbitrary kernel write or control-flow hijack, not merely a clean crash.
- A:HAvailability impact, high
- Invoking might_sleep operations from softirq or RCU-critical sections reliably causes kernel BUG/oops/panic ("sleeping function called from invalid context"), producing complete host denial of service.