CVE-2026-74638
In the Linux kernel, the following vulnerability has been resolved:
drm/v3d: Serialize the scheduler timeout handlers
V3D exposes several independent hardware queues (BIN, RENDER, TFU and CSD) but has only a single, global reset. A timeout on any one queue therefore has to stop, reset and restart the schedulers of every other queue as well. That makes concurrent timeout handlers unsafe.
`reset_lock` was never able to make them safe, as a driver-side lock can only cover the driver's &drm_sched_backend_ops.timedout_job callback.
Leer descripción completaMostrar menos
The scheduler handles the timed out job and its pending list around that callback, outside of the driver's control, so a global reset triggered by one queue can still interfere with another queue that is in the middle of handling a timeout of its own.
Consequently, if a reset happens in the CSD queue while a CL-intensive application is running, the global reset stops and restarts the CL queue's scheduler while that queue is handling a timeout of its own. As drm_sched_stop() and drm_sched_start() subtract and add the credits of every job sitting on the pending list of the scheduler they are called on, and as the CL queue's handler concurrently takes its job off that same list and puts it back, the stop and the start no longer see the same set of jobs. The CL queue is left with more credits in flight than its limit:
From that point on, the credit count of the CL queue is broken, causing a complete GPU hang and UI freeze.
The DRM scheduler already provides a mechanism to serialize the timeout handlers of different schedulers: an ordered workqueue passed as drm_sched_init()'s @timeout_wq parameter. By default, each scheduler queues its timeout work on the system workqueue, which runs the handlers concurrently. Give all of the queues a shared ordered workqueue instead, as recommended by the DRM scheduler documentation for hardware that has distinct queues but resets globally.
Detalles técnicos trazas, registros y código del informe original
[ 327.302739] ------------[ cut here ]------------ [ 327.302744] WARNING: CPU: 2 PID: 43 at drivers/gpu/drm/scheduler/sched_main.c:102 drm_sched_run_job_work+0x238/0x4d0 [gpu_sched] [ 327.302884] CPU: 2 UID: 0 PID: 43 Comm: kworker/u16:1 Not tainted 6.18.39-v8-16k+ #3 PREEMPT [ 327.302889] Hardware name: Raspberry Pi 5 Model B Rev 1.0 (DT) [ 327.302893] Workqueue: v3d_bin drm_sched_run_job_work [gpu_sched] [ 327.302984] Call trace: [ 327.302987] drm_sched_run_job_work+0x238/0x4d0 [gpu_sched] (P) [ 327.302997] process_scheduled_works+0x180/0x3d0 [ 327.303010] worker_thread+0x268/0x3e8 [ 327.303016] kthread+0x140/0x250 [ 327.303022] ret_from_fork+0x10/0x20 [ 327.303031] ---[ end trace 0000000000000000 ]---
CVSS
NVD no ha asignado puntuación CVSS a esta CVE (habitual desde el cambio de política de abril de 2026).
Probabilidad de explotación (EPSS)
- Probabilidad de explotación en los próximos 30 días: 0.21%
- Percentil entre todas las CVEs puntuadas: 10
- Fecha de la puntuación: 5/10/2026
EPSS (Exploit Prediction Scoring System, de FIRST) estima la probabilidad de que una vulnerabilidad sea explotada en 30 días. Complementa a CVSS (impacto) y a CISA KEV (explotación confirmada).
Tecnologías afectadas (1)
⚠ Inferidas por IA a partir de la descripción — NVD aún no ha analizado esta CVE; no son CPE verificados.
Referencias
JSON original (NVD)
Mostrar
{
"id": "CVE-2026-74638",
"cveTags": [],
"metrics": {},
"affected": [
{
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
"affectedData": [
{
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"vendor": "Linux",
"product": "Linux",
"versions": [
{
"status": "affected",
"version": "35e4079bf1a2570abffce6ababa631afcf8ea0e5",
"lessThan": "5884851a096d8afcdf91f0e542bac193035183a6",
"versionType": "git"
},
{
"status": "affected",
"version": "35e4079bf1a2570abffce6ababa631afcf8ea0e5",
"lessThan": "c22a45817b9c92aa0391db60e2ed467c7e6027d7",
"versionType": "git"
},
{
"status": "affected",
"version": "35e4079bf1a2570abffce6ababa631afcf8ea0e5",
"lessThan": "4da94744707b27a3ae1197bdd7127da4505dc5b1",
"versionType": "git"
},
{
"status": "affected",
"version": "5235b56b7e5449d990d21d78723b1a5e7bb5738e",
"versionType": "git"
},
{
"status": "affected",
"version": "12125f7d9c15e6d8ac91d10373b2db2f17dcf767",
"versionType": "git"
},
{
"status": "affected",
"version": "a5f162727b91e480656da1876247a91f651f76de",
"versionType": "git"
},
{
"status": "affected",
"version": "422a8b10ba42097a704d6909ada2956f880246f2",
"versionType": "git"
},
{
"status": "affected",
"version": "6.1.139",
"lessThan": "6.2",
"versionType": "semver"
},
{
"status": "affected",
"version": "6.6.91",
"lessThan": "6.7",
"versionType": "semver"
},
{
"status": "affected",
"version": "6.12.29",
"lessThan": "6.13",
"versionType": "semver"
},
{
"status": "affected",
"version": "6.14.7",
"lessThan": "6.15",
"versionType": "semver"
}
],
"programFiles": [
"drivers/gpu/drm/v3d/v3d_drv.h",
"drivers/gpu/drm/v3d/v3d_sched.c"
],
"defaultStatus": "unaffected"
},
{
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"vendor": "Linux",
"product": "Linux",
"versions": [
{
"status": "affected",
"version": "6.15"
},
{
"status": "unaffected",
"version": "0",
"lessThan": "6.15",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "6.18.45",
"versionType": "semver",
"lessThanOrEqual": "6.18.*"
},
{
"status": "unaffected",
"version": "7.1.9",
"versionType": "semver",
"lessThanOrEqual": "7.1.*"
},
{
"status": "unaffected",
"version": "7.2",
"versionType": "original_commit_for_fix",
"lessThanOrEqual": "*"
}
],
"programFiles": [
"drivers/gpu/drm/v3d/v3d_drv.h",
"drivers/gpu/drm/v3d/v3d_sched.c"
],
"defaultStatus": "affected"
}
]
}
],
"published": "2026-08-22T16:16:37.007",
"references": [
{
"url": "https://git.kernel.org/stable/c/4da94744707b27a3ae1197bdd7127da4505dc5b1",
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
},
{
"url": "https://git.kernel.org/stable/c/5884851a096d8afcdf91f0e542bac193035183a6",
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
},
{
"url": "https://git.kernel.org/stable/c/c22a45817b9c92aa0391db60e2ed467c7e6027d7",
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
}
],
"vulnStatus": "Received",
"descriptions": [
{
"lang": "en",
"value": "In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/v3d: Serialize the scheduler timeout handlers\n\nV3D exposes several independent hardware queues (BIN, RENDER, TFU and\nCSD) but has only a single, global reset. A timeout on any one queue\ntherefore has to stop, reset and restart the schedulers of every other\nqueue as well. That makes concurrent timeout handlers unsafe.\n\n`reset_lock` was never able to make them safe, as a driver-side lock can\nonly cover the driver's &drm_sched_backend_ops.timedout_job callback.\nThe scheduler handles the timed out job and its pending list around that\ncallback, outside of the driver's control, so a global reset triggered\nby one queue can still interfere with another queue that is in the\nmiddle of handling a timeout of its own.\n\nConsequently, if a reset happens in the CSD queue while a CL-intensive\napplication is running, the global reset stops and restarts the CL\nqueue's scheduler while that queue is handling a timeout of its own. As\ndrm_sched_stop() and drm_sched_start() subtract and add the credits of\nevery job sitting on the pending list of the scheduler they are called\non, and as the CL queue's handler concurrently takes its job off that\nsame list and puts it back, the stop and the start no longer see the\nsame set of jobs. The CL queue is left with more credits in flight than\nits limit:\n\n[ 327.302739] ------------[ cut here ]------------\n[ 327.302744] WARNING: CPU: 2 PID: 43 at drivers/gpu/drm/scheduler/sched_main.c:102 drm_sched_run_job_work+0x238/0x4d0 [gpu_sched]\n[ 327.302884] CPU: 2 UID: 0 PID: 43 Comm: kworker/u16:1 Not tainted 6.18.39-v8-16k+ #3 PREEMPT\n[ 327.302889] Hardware name: Raspberry Pi 5 Model B Rev 1.0 (DT)\n[ 327.302893] Workqueue: v3d_bin drm_sched_run_job_work [gpu_sched]\n[ 327.302984] Call trace:\n[ 327.302987] drm_sched_run_job_work+0x238/0x4d0 [gpu_sched] (P)\n[ 327.302997] process_scheduled_works+0x180/0x3d0\n[ 327.303010] worker_thread+0x268/0x3e8\n[ 327.303016] kthread+0x140/0x250\n[ 327.303022] ret_from_fork+0x10/0x20\n[ 327.303031] ---[ end trace 0000000000000000 ]---\n\nFrom that point on, the credit count of the CL queue is broken, causing\na complete GPU hang and UI freeze.\n\nThe DRM scheduler already provides a mechanism to serialize the timeout\nhandlers of different schedulers: an ordered workqueue passed as\ndrm_sched_init()'s @timeout_wq parameter. By default, each scheduler\nqueues its timeout work on the system workqueue, which runs the handlers\nconcurrently. Give all of the queues a shared ordered workqueue instead,\nas recommended by the DRM scheduler documentation for hardware that has\ndistinct queues but resets globally."
}
],
"lastModified": "2026-08-25T06:18:43.493",
"sourceIdentifier": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
}