Look up another CVE · open CVE-2026-90358 in the interactive tracker
CVE-2026-90358
bpf, x86: Fix trampoline stack size for 128-bit arguments
Published 2026-09-17 (31 days ago) by the Linux kernel CNA. Subsystem arch/x86. Debian urgency: not yet assigned.
What to do, per Debian release
Debian's own status for CVE-2026-90358 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 | vulnerable | 6.1.190-1 | No fix published anywhere yet, neither in Debian nor upstream for the series bookworm tracks. Watch the Debian tracker page for this CVE. Advisories for bookworm have historically landed a median of 23 days after publication. |
| Debian 13trixie | fixedDSA-6528-1 · 2026-09-29 · text | 6.12.111-1 | Update the kernel and reboot. DSA-6528-1 shipped the fix on 2026-09-29, in 6.12.111-1. A running kernel keeps the old code until the machine restarts. |
| Debian 14forky | fixed | 7.2.6-1 | Update the kernel and reboot. Fixed in 7.2.6-1. A running kernel keeps the old code until the machine restarts. |
| Debian unstablesid | fixed | 7.2.6-1 | Update the kernel and reboot. Fixed in 7.2.6-1. A running kernel keeps the old code until the machine restarts. |
| Debian 12bookworm · linux-6.12 | fixedDLA-4817-1 · 2026-10-04 · text | 6.12.111-1~deb12u1 | Update the kernel and reboot. DLA-4817-1 shipped the fix on 2026-10-04, in 6.12.111-1~deb12u1. A running kernel keeps the old code until the machine restarts. |
The vulnerability
What the bug is
btf_distill_func_proto() accepts a function argument up to 16 bytes, so a 128-bit scalar such as __int128 reaches the x86 trampoline with arg_size == 16. But the current implementation assumes an __int128 argument only needs one register, so the register save area is under-allocated and save_args() overwrites adjacent stack slots. Compute the register count from arg_size for all arguments to fix it.
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.
arch/x86/net/bpf_jit_comp.c
- What this is
- BPF JIT for X86 64-BITMAINTAINERS section matched on arch/x86/net/ · bpf@vger.kernel.org · Supported
Check it on the machine
grep -w CONFIG_BPF_JIT /boot/config-$(uname -r)Reads the booted kernel's own configuration, on the machine and nowhere else. =y is built in, =m is a module, and an is not set line means the code is not there.
Show whether Debian builds this, per release
Debian builds this, but not on every flavour. At least one release below builds the code on some of its kernel flavours and not on others, so the answer depends on the architecture and flavour installed.
- bookwormy
- trixie, forky, sidYnot built on s390x/s390x
- bookworm · linux-6.12y
- y
- built into the kernel image on every flavour - only a patch removes it
- Y
- built in where it is built at all; some flavours do not build it
- Build switch
- CONFIG_BPF_JIT
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 kbuild Makefiles Debian linux-config packages
Signals
- Published
- 2026-09-17 (31 days ago)source
- CVSS
- 7.8 high
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:Hsource recompute it - Attack vector
- local (AV:L)source
- EPSS
- 0.16%, higher than 5.2% of all CVEssource
- CISA KEV
- not listedthe KEV catalogue
- Debian urgency
- not yet assignedsource
- Subsystem
- arch/x86
Upstream fixes
The versions the CNA records as introducing and fixing this, and the commit for each. source
| Introduced | Fixed in | Commit |
|---|---|---|
| 6.1 | 6.12.110 | 4aed2346fa9b |
| 6.1 | 6.18.52 | 40f3bfbbb222 |
| 6.1 | 7.2.6 | a6d7d7a253c5 |
| 6.1 | 7.3 | 814cba835ef6 |
Debian advisories
| Advisory | Date | Shipped to | Debian |
|---|---|---|---|
| DLA-4817-1 | 2026-10-04 | bookworm 6.12.111-1~deb12u1 | tracker |
| DSA-6528-1 | 2026-09-29 | trixie 6.12.111-1 | tracker |
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
- The under-allocation is in __arch_prepare_bpf_trampoline()/save_args() in arch/x86/net/bpf_jit_comp.c and is reached only when a local process issues bpf(BPF_PROG_LOAD) for BPF_PROG_TYPE_TRACING fentry/fexit and bpf(BPF_LINK_CREATE) against a BTF function with a 16-byte integer argument. BPF is a local syscall vector, not a network protocol parser.
- AC:LAttack complexity, low
- The attacker selects an attach target whose BTF has arg_size==16 without BTF_FMODEL_STRUCT_ARG (e.g. bpf_testmod_test_int128_arg). __arch_prepare_bpf_trampoline() then under-counts nr_regs by one slot per such argument and save_args() deterministically overwrites the next stack slot; no race or uninfluenced layout is required.
- PR:LPrivileges required, low
- bpf_prog_load() rejects BPF_PROG_TYPE_TRACING unless bpf_token_capable(CAP_BPF) and, via is_perfmon_prog_type(), bpf_token_capable(CAP_PERFMON). Those capabilities are delegable with BPF tokens and are scored Low, not init-namespace root, consistent with other BPF trampoline CVEs.
- UI:NUser interaction, none
- The attacker loads the fentry/fexit program, attaches it with BPF_LINK_CREATE, and invokes the target themselves (for bpf_testmod, a read of the module sysfs test file that calls bpf_testmod_test_int128_arg). No separate victim mount or click is required.
- S:UScope, unchanged
- save_args() overwrites the trampoline's own saved RBP and adjacent slots in the host kernel stack. That is in-kernel memory corruption on the same authority, not a KVM/Xen guest-to-host, IOMMU, or sandbox boundary crossing.
- C:HConfidentiality impact, high
- The extra 8-byte store of the __int128's second register (RSI) lands in adjacent trampoline slots. On an fentry trampoline without a return-value slot that is saved RBP, so leave pivots RSP through argument contents and subsequent pops/unwinds can disclose kernel stack/memory.
- I:HIntegrity impact, high
- The same OOB store overwrites saved RBP (and with two 16-byte scalars, the return address at RBP+8). The trampoline epilogue then executes leave; ret on a corrupted frame, which is kernel stack memory corruption exploitable for a stack pivot and control-flow hijack.
- A:HAvailability impact, high
- Overwriting saved RBP with the high half of a small __int128 (0 for bpf_testmod_test_int128_arg((__int128)1,...)) makes leave load a bogus frame pointer, so the following pop/ret oopses or panics the kernel.