IM
IronMonkey Threat Research

CVE-2024-26958 HIGH

Published: 2024-05-01 | Last Modified: 2026-08-04 | Status: Modified

Description

In the Linux kernel, the following vulnerability has been resolved: nfs: fix UAF in direct writes In production we have been hitting the following warning consistently ------------[ cut here ]------------ refcount_t: underflow; use-after-free. WARNING: CPU: 17 PID: 1800359 at lib/refcount.c:28 refcount_warn_saturate+0x9c/0xe0 Workqueue: nfsiod nfs_direct_write_schedule_work [nfs] RIP: 0010:refcount_warn_saturate+0x9c/0xe0 PKRU: 55555554 Call Trace: <TASK> ? __warn+0x9f/0x130 ? refcount_warn_saturate+0x9c/0xe0 ? report_bug+0xcc/0x150 ? handle_bug+0x3d/0x70 ? exc_invalid_op+0x16/0x40 ? asm_exc_invalid_op+0x16/0x20 ? refcount_warn_saturate+0x9c/0xe0 nfs_direct_write_schedule_work+0x237/0x250 [nfs] process_one_work+0x12f/0x4a0 worker_thread+0x14e/0x3b0 ? ZSTD_getCParams_internal+0x220/0x220 kthread+0xdc/0x120 ? __btf_name_valid+0xa0/0xa0 ret_from_fork+0x1f/0x30 This is because we're completing the nfs_direct_request twice in a row. The source of this is when we have our commit requests to submit, we process them and send them off, and then in the completion path for the commit requests we have if (nfs_commit_end(cinfo.mds)) nfs_direct_write_complete(dreq); However since we're submitting asynchronous requests we sometimes have one that completes before we submit the next one, so we end up calling complete on the nfs_direct_request twice. The only other place we use nfs_generic_commit_list() is in __nfs_commit_inode, which wraps this call in a nfs_commit_begin(); nfs_commit_end(); Which is a common pattern for this style of completion handling, one that is also repeated in the direct code with get_dreq()/put_dreq() calls around where we process events as well as in the completion paths. Fix this by using the same pattern for the commit requests. Before with my 200 node rocksdb stress running this warning would pop every 10ish minutes. With my patch the stress test has been running for several hours without popping.

Additional Descriptions (1)

En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: nfs: corrige UAF en escrituras directas En producción hemos estado recibiendo la siguiente advertencia constantemente ------------[ cortar aquí ]----- ------- refcount_t: desbordamiento insuficiente; use-after-free. ADVERTENCIA: CPU: 17 PID: 1800359 en lib/refcount.c:28 refcount_warn_saturate+0x9c/0xe0 Cola de trabajo: nfsiod nfs_direct_write_schedule_work [nfs] RIP: 0010:refcount_warn_saturate+0x9c/0xe0 PKRU: 55555554 Llamada Seguimiento: ? __advertir+0x9f/0x130 ? refcount_warn_saturate+0x9c/0xe0? report_bug+0xcc/0x150? handle_bug+0x3d/0x70? exc_invalid_op+0x16/0x40? asm_exc_invalid_op+0x16/0x20? refcount_warn_saturate+0x9c/0xe0 nfs_direct_write_schedule_work+0x237/0x250 [nfs] Process_one_work+0x12f/0x4a0 trabajador_thread+0x14e/0x3b0? ZSTD_getCParams_internal+0x220/0x220 kthread+0xdc/0x120? __btf_name_valid+0xa0/0xa0 ret_from_fork+0x1f/0x30 Esto se debe a que estamos completando nfs_direct_request dos veces seguidas. La fuente de esto es cuando tenemos que enviar nuestras solicitudes de confirmación, las procesamos y las enviamos, y luego en la ruta de finalización para las solicitudes de confirmación tenemos if (nfs_commit_end(cinfo.mds)) nfs_direct_write_complete(dreq); Sin embargo, dado que enviamos solicitudes asincrónicas, a veces tenemos una que se completa antes de enviar la siguiente, por lo que terminamos llamando a complete en nfs_direct_request dos veces. El único otro lugar donde usamos nfs_generic_commit_list() es en __nfs_commit_inode, que envuelve esta llamada en nfs_commit_begin(); nfs_commit_end(); Este es un patrón común para este estilo de manejo de finalización, uno que también se repite en el código directo con llamadas get_dreq()/put_dreq() en el lugar donde procesamos eventos, así como en las rutas de finalización. Solucione este problema utilizando el mismo patrón para las solicitudes de confirmación. Antes, con mi estrés de rocksdb de 200 nodos ejecutándose, esta advertencia aparecía cada 10 minutos. Con mi parche, la prueba de esfuerzo ha estado funcionando durante varias horas sin aparecer.

CVSS Metrics

Base Score: 7.8 (HIGH)

CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

Attack VectorLOCAL
Attack ComplexityLOW
Privileges RequiredLOW
User InteractionNONE
ScopeUNCHANGED
Confidentiality ImpactHIGH
Integrity ImpactHIGH
Availability ImpactHIGH

Source: [email protected]

Type: Primary

Exploitability Score: 1.8

Impact Score: 5.9

Weaknesses

Source Type Description
[email protected] Primary
en CWE-416

Affected Products

Vendor Product Version Update Type
linux linux_kernel * <built-in method update of dict object at 0x7fd3686cb340> Operating System
linux linux_kernel * <built-in method update of dict object at 0x7fd368ee4e80> Operating System
linux linux_kernel * <built-in method update of dict object at 0x7fd3855d8b00> Operating System
linux linux_kernel * <built-in method update of dict object at 0x7fd3105f95c0> Operating System
linux linux_kernel * <built-in method update of dict object at 0x7fd3686ca280> Operating System
linux linux_kernel * <built-in method update of dict object at 0x7fd368ee4dc0> Operating System
debian debian_linux 10.0 <built-in method update of dict object at 0x7fd3686d1780> Operating System

Affected Configurations

Operator: OR

Vulnerable CPE
Yes cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*
Yes cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*
Yes cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*
Yes cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*
Yes cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*
Yes cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*

Operator: OR

Vulnerable CPE
Yes cpe:2.3:o:debian:debian_linux:10.0:*:*:*:*:*:*:*

References