Look up another CVE · open CVE-2026-97433 in the interactive tracker
CVE-2026-97433
nvme: validate FDP configuration descriptor sizes
Published 2026-09-24 (24 days ago) by the Linux kernel CNA. Subsystem drivers/nvme. Debian urgency: unimportant.
What to do, per Debian release
Debian's own status for CVE-2026-97433 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 | not affected | 6.1.190-1 | Nothing to do. bookworm never shipped the vulnerable code. |
| Debian 13trixie | not affected | 6.12.113-1 | Nothing to do. trixie never shipped the vulnerable code. |
| 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. |
The vulnerability
What the bug is
Validate descriptor sizes while walking the FDP configurations log so dsze == 0 or a descriptor past the log end cannot cause unbounded iteration or reads past the buffer.
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.
drivers/nvme/host/core.c
- What this is
- NVM EXPRESS DRIVERMAINTAINERS section matched on drivers/nvme/host/ · linux-nvme@lists.infradead.org · Supported
- Kernel module
- nvme_core
Check it on the machine
grep -w CONFIG_NVME_CORE /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.
lsmod | grep -w nvme_corePrints a line if that module is loaded right now, and nothing if it is not. One that is not loaded can still be loaded later by anything that needs it.
Show whether Debian builds this, per release
Debian builds this code. Every release below compiles the affected file into the kernels it ships.
- bookworm, trixie, forky, sidb
- b
- built in on some flavours, a module on others
- Build switch
- CONFIG_NVME_CORE
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
- 2026-09-24 (24 days ago)source
- CVSS
- 8.2 high
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:Hsource recompute it - Attack vector
- network (AV:N)source
- EPSS
- 0.32%, higher than 23.3% of all CVEssource
- CISA KEV
- not listedthe KEV catalogue
- Debian urgency
- unimportantsource
- Subsystem
- drivers/nvme
Upstream fixes
The versions the CNA records as introducing and fixing this, and the commit for each. source
| Introduced | Fixed in | Commit |
|---|---|---|
| 6.16 | 6.18.53 | 97efd7a41aac |
| 6.16 | 7.2 | 0ef4daa6534a |
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:NAttack vector, network
- nvme_query_fdp_granularity() walks nvme_fdp_config_desc.dsze from the Get Log Page LID 0x20 (NVME_LOG_FDP_CONFIGS) buffer filled by nvme_get_log_lsi(); on nvme-tcp that payload is C2HData copied by nvme_tcp_recv_data() from the remote NVMe-oF target (TCP 4420), so the malformed descriptor sizes arrive on the wire.
- AC:LAttack complexity, low
- The peer sets Identify ctratt NVME_CTRL_ATTR_FDPS, Get Features FID 0x1d (FDPE and fdpcidx), and log fields sze/dsze, so nvme_query_fdp_granularity() can skip with a crafted dsze to park desc just below end and then load nrg/runs past the kvmalloc object with no race.
- PR:NPrivileges required, none
- The attacker is the NVMe-oF controller. nvme_scan_work() submits that Get Log Page on the existing admin_q; DH-HMAC-CHAP and TLS are optional, and no host uid, capability, or /dev/nvme-fabrics write is required once the session exists.
- UI:NUser interaction, none
- nvme_start_ctrl() queues ctrl->scan_work and nvme_handle_aen_notice() requeues it on NVME_AER_NOTICE_NS_CHANGED, so nvme_scan_ns()->nvme_update_ns_info_block()->nvme_query_fdp_info()->nvme_query_fdp_granularity() runs on nvme_wq without a further victim mount or click.
- S:UScope, unchanged
- The over-read is of the host kvmalloc FDP configs log and the info->runs store inside nvme_query_fdp_granularity(); it does not cross a KVM/Xen, IOMMU, or other distinct security authority.
- C:LConfidentiality impact, low
- After the dsze walk, nvme_query_fdp_granularity() only does le32_to_cpu(desc->nrg) and le64_to_cpu(desc->runs) (offsets 4 and 16). With desc at end-1 that is a strictly bounded 12-byte load past the log object, a small heap leak rather than an arbitrary read.
- I:NIntegrity impact, none
- Those fields are stored to info->runs and min()'d into write_stream_granularity in nvme_update_ns_info_block(); the dsze walk performs no out-of-bounds write, use-after-free, or control-flow hijack.
- A:HAvailability impact, high
- hdr.sze is attacker-chosen (capped at PAGE_SIZE*MAX_ORDER_NR_PAGES) for kvmalloc(), and desc->nrg/desc->runs are then loaded without ensuring the selected descriptor fits; that over-read past the allocation can oops when the walk parks desc at the buffer end.