nday
CVE-2026-45837

bpf: Fix use-after-free in arena_vm_close on fork

Linux / Linux

Fecha de divulgación: . nday aún no tiene constancia de una corrección; eso no significa que no exista parche. Ninguna de las cuatro fuentes de PoC la recoge. No figura en el KEV de CISA.

tags weakness typeuse-after-free accesslocalno user interaction productLinux kernel product typeoperating system
why these tags

nday derives these tags by fixed rules from the facts below; NVD, the CNA and other sources do not assign them. Vocabulary and rules: ~/tags.

use-after-free
CWE-416, stated by NVD's analysts
local
AV:L in CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H (CVSS 3.1) from NVD's analysts
no user interaction
UI:N in CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H (CVSS 3.1) from NVD's analysts
Linux kernel
affected products in the CVE record: Linux · LinuxCPE names NVD's analysts give as vulnerable: o:linux:linux_kernel
operating system
affected products in the CVE record: Linux · LinuxCPE names NVD's analysts give as vulnerable: o:linux:linux_kernelthe product type of Linux kernel in nday's product registry
etapa
DIS
divulgación
cvss
sin puntuación aún

Ciclo de vida

Eje x en UTC; punto = minuto exacto, barra ancha = solo fecha, hueco = desconocida o inferida.
fecha desconocida
DISdivulgada
05-27T09:24Z
PATparcheada
sin registro aún
ANAanalizada
sin registro aún
POCPoC
sin registro aún
ITWexplotada
sin registro aún
KEVen KEV
sin registro aún

Descripción

In the Linux kernel, the following vulnerability has been resolved:

bpf: Fix use-after-free in arena_vm_close on fork

arena_vm_open() only bumps vml->mmap_count but never registers the child VMA in arena->vma_list. The vml->vma always points at the parent VMA, so after parent munmap the pointer dangles. If the child then calls bpf_arena_free_pages(), zap_pages() reads the stale vml->vma triggering use-after-free.

Fix this by preventing the arena VMA from being inherited across fork with VM_DONTCOPY, and preventing VMA splits via the may_split callback.

Also reject mremap with a .mremap callback returning -EINVAL. A same-size mremap(MREMAP_FIXED) on the full arena VMA reaches copy_vma() through the following path:

check_prep_vma()       - returns 0 early: new_len == old_len
                         skips VM_DONTEXPAND check
prep_move_vma()        - vm_start == old_addr and
                         vm_end == old_addr + old_len
                         so may_split is never called
move_vma()
  copy_vma_and_data()
    copy_vma()
      vm_area_dup()    - copies vm_private_data (vml pointer)
      vm_ops->open()   - bumps vml->mmap_count
    vm_ops->mremap()   - returns -EINVAL, rollback unmaps new VMA

The refcount ensures the rollback's arena_vm_close does not free the vml shared with the original VMA.

Original text of the CVE record (cvelist), shown as published.

Registro de eventos

+DISdivulgación

Linux publicó el registro CVE.

fuente cvelist · nday , importación masiva · en bruto
{
  "id": 426008,
  "type": "disclosed",
  "source": "cvelist",
  "url": "https://www.cve.org/CVERecord?id=CVE-2026-45837",
  "occurred_at": "2026-05-27T09:24:32.833Z",
  "observed_at": "2026-11-13T05:39:50.021609Z",
  "is_backfill": true,
  "dedupe_key": "",
  "detail": {
    "state": "PUBLISHED",
    "date_published": "2026-05-27T09:24:32.833Z",
    "assigner_short_name": "Linux"
  }
}

Report an error

report an error or omission

Something on this page wrong or missing? Tell the editors; if they accept it, this page's revision log says so publicly. Upstream data is not changed.

Debate de investigadores

0 investigadores lo mencionaron en los últimos 7 días

Publicaciones en X de investigadores de la lista de nday que mencionan este CVE, tal como se publicaron. nday no las verifica.

nday aún no ha recopilado publicaciones en X de investigadores de su lista sobre este CVE.

Discussion

0 comments

No comments yet.