Look up another CVE · open CVE-2026-74383 in the interactive tracker
CVE-2026-74383
nvme-pci: fix out-of-bounds access in nvme_setup_descriptor_pools
Published 2026-08-15 (2 months ago) by the Linux kernel CNA. Subsystem drivers/nvme. Debian urgency: unimportant.
What to do, per Debian release
Debian's own status for CVE-2026-74383 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 | 7.1.5-1 | Update the kernel and reboot. Fixed in 7.1.5-1. A running kernel keeps the old code until the machine restarts. |
| Debian unstablesid | fixed | 7.1.5-1 | Update the kernel and reboot. Fixed in 7.1.5-1. A running kernel keeps the old code until the machine restarts. |
The vulnerability
What the bug is
nvme_setup_descriptor_pools() indexes dev->descriptor_pools[] using the numa_node forwarded from hctx->numa_node by its single caller, nvme_init_hctx_common(). On a non-NUMA kernel hctx->numa_node is NUMA_NO_NODE (-1). Because the parameter was declared 'unsigned', the value becomes UINT_MAX and the index walks off the array (sized to nr_node_ids), faulting during nvme_alloc_ns() and leaving the namespace without a /dev node. Reproduces on any NVMe controller probed by a CONFIG_NUMA=n kernel: BUG: unable to handle page fault for address: ffff889101603d38 RIP: 0010:nvme_init_hctx_common+0x5a/0x190 [nvme] Call Trace: nvme_init_hctx+0x10/0x20 [nvme] nvme_alloc_ns+0x9e/0xa10 [nvme_core] nvme_scan_ns+0x301/0x3b0 [nvme_core] nvme_scan_ns_async+0x23/0x30 [nvme_core] Switch the parameter to int and fall back to node 0 when it is NUMA_NO_NODE; node 0 is always present.
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/pci.c
- What this is
- NVM EXPRESS DRIVERMAINTAINERS section matched on drivers/nvme/host/ · linux-nvme@lists.infradead.org · Supported
- Kernel module
- nvme
Check it on the machine
grep -w CONFIG_BLK_DEV_NVME /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 nvmePrints 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, 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.
- bookwormb
- trixieBnot built on armel/rpi
- forky, sidb
- b
- built in on some flavours, a module on others
- B
- built in on some, a module on others, absent on the rest
- Build switch
- CONFIG_BLK_DEV_NVME
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
- 2026-08-15 (2 months ago)source
- CVSS
- 8.4 high
CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:Hsource recompute it - Attack vector
- local (AV:L)source
- EPSS
- 0.19%, higher than 8.4% 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 | 6.18.40 | 3e8aed5edaee |
| 6.16 | 7.1.5 | fa9b6accd1ad |
| 6.16 | 7.2 | a192b8cfa447 |
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 flaw is in the nvme-pci host driver and is reached during local PCIe NVMe controller probe and namespace scan (nvme_probe/nvme_scan_work→nvme_alloc_ns→blk_mq_alloc_disk→nvme_init_hctx_common), including rescan via NVME_IOCTL_RESCAN/sysfs and AER NS_CHANGED events from a locally attached device; there is no network-facing entry point.
- AC:LAttack complexity, low
- On CONFIG_NUMA=n kernels, hctx->numa_node is always NUMA_NO_NODE, so the unsigned wrap to UINT_MAX and subsequent out-of-bounds access occur deterministically the first time any namespace hardware context is initialized for a probed NVMe controller, without races or attacker-influenced memory layout.
- PR:NPrivileges required, none
- Exploitation requires no host OS credentials: any locally attached PCIe NVMe controller (boot-present, hot-plugged, supply-chain, or kiosk/shared-workstation storage) triggers the vulnerable scan during probe or when device firmware delivers a namespace-changed async event, without CAP_SYS_ADMIN or other privileges.
- UI:NUser interaction, none
- Triggering the bug does not require any victim user action such as mounting a filesystem or opening a device node; namespace scanning runs automatically from driver probe, controller rescan, or device-generated AER notifications handled entirely in kernel workqueues.
- S:UScope, unchanged
- Impact is confined to kernel memory corruption and crash within the host kernel NVMe/PCI driver security domain; it does not cross a VM, container, or IOMMU boundary to compromise a separate security authority.
- C:HConfidentiality impact, high
- The vulnerable code performs an out-of-bounds read of descriptor_pools[UINT_MAX] (checking pools->small) far beyond the nr_node_ids-sized allocation, constituting an unbounded kernel memory read that can disclose adjacent kernel data if the access does not immediately fault.
- I:HIntegrity impact, high
- The same out-of-bounds index would be used for dma_pool_create_node writes to pools->large and pools->small, giving an out-of-bounds kernel write primitive; even when the fault occurs on the initial read, this is exploitable-class memory corruption rather than a benign NULL dereference.
- A:HAvailability impact, high
- The reported reproducer causes a kernel page fault in nvme_init_hctx_common during nvme_alloc_ns, producing an oops that aborts namespace creation and can panic the system, denying storage availability on CONFIG_NUMA=n hosts with NVMe controllers.