CVE-2026-23435
In the Linux kernel, the following vulnerability has been resolved:
perf/x86: Move event pointer setup earlier in x86_pmu_enable()
A production AMD EPYC system crashed with a NULL pointer dereference in the PMU NMI handler:
The faulting instruction is `cmpq $0x0, 0x198(%rdi)` with RDI=0, corresponding to the `if (unlikely(!hwc->event_base))` check in x86_perf_event_update() where hwc = &event->hw and event is NULL.
drgn inspection of the vmcore on CPU 106 showed a mismatch between cpuc->active_mask and cpuc->events[]:
The event that should occupy events[2] was found in event_list[2] with hw.idx=2 and hw.state=0x0, confirming x86_pmu_start() had run (which clears hw.state and sets active_mask) but events[2] was never populated.
Leer descripción completaMostrar menos
Another event (event_list[0]) had hw.state=0x7 (STOPPED|UPTODATE|ARCH), showing it was stopped when the PMU rescheduled events, confirming the throttle-then-reschedule sequence occurred.
The root cause is commit 7e772a93eb61 ("perf/x86: Fix NULL event access and potential PEBS record loss") which moved the cpuc->events[idx] assignment out of x86_pmu_start() and into step 2 of x86_pmu_enable(), after the PERF_HES_ARCH check. This broke any path that calls pmu->start() without going through x86_pmu_enable() -- specifically the unthrottle path:
The race sequence is:
Move the cpuc->events[hwc->idx] assignment in x86_pmu_enable() to before the PERF_HES_ARCH check, so that events[] is populated even for events that are not immediately started. This ensures the unthrottle path via pmu->start() always finds a valid event pointer.
Detalles técnicos trazas, registros y código del informe original
BUG: kernel NULL pointer dereference, address: 0000000000000198
RIP: x86_perf_event_update+0xc/0xa0
Call Trace:
<NMI>
amd_pmu_v2_handle_irq+0x1a6/0x390
perf_event_nmi_handler+0x24/0x40
active_mask: 0x1e (bits 1, 2, 3, 4)
events[1]: 0xff1100136cbd4f38 (valid)
events[2]: 0x0 (NULL, but active_mask bit 2 set)
events[3]: 0xff1100076fd2cf38 (valid)
events[4]: 0xff1100079e990a90 (valid)
perf_adjust_freq_unthr_events()
-> perf_event_unthrottle_group()
-> perf_event_unthrottle()
-> event->pmu->start(event, 0)
-> x86_pmu_start() // sets active_mask but not events[]
1. A group of perf events overflows, triggering group throttle via
perf_event_throttle_group(). All events are stopped: active_mask
bits cleared, events[] preserved (x86_pmu_stop no longer clears
events[] after commit 7e772a93eb61).
2. While still throttled (PERF_HES_STOPPED), x86_pmu_enable() runs
due to other scheduling activity. Stopped events that need to
move counters get PERF_HES_ARCH set and events[old_idx] cleared.
In step 2 of x86_pmu_enable(), PERF_HES_ARCH causes these events
to be skipped -- events[new_idx] is never set.
3. The timer tick unthrottles the group via pmu->start(). Since
commit 7e772a93eb61 removed the events[] assignment from
x86_pmu_start(), active_mask[new_idx] is set but events[new_idx]
remains NULL.
4. A PMC overflow NMI fires. The handler iterates active counters,
finds active_mask[2] set, reads events[2] which is NULL, and
crashes dereferencing it.CVSS
- Versión: 3.1
- Vector: CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
- Puntuación base: 5.5
Probabilidad de explotación (EPSS)
- Probabilidad de explotación en los próximos 30 días: 0.12%
- Percentil entre todas las CVEs puntuadas: 2
- Fecha de la puntuación: 6/10/2026
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)
CWE
- CWE-476
Referencias
JSON original (NVD)
Mostrar
{
"id": "CVE-2026-23435",
"cveTags": [],
"metrics": {
"cvssMetricV31": [
{
"type": "Primary",
"source": "nvd@nist.gov",
"cvssData": {
"scope": "UNCHANGED",
"version": "3.1",
"baseScore": 5.5,
"attackVector": "LOCAL",
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
"integrityImpact": "NONE",
"userInteraction": "NONE",
"attackComplexity": "LOW",
"availabilityImpact": "HIGH",
"privilegesRequired": "LOW",
"confidentialityImpact": "NONE"
},
"impactScore": 3.6,
"exploitabilityScore": 1.8
}
]
},
"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": "6b089028bff1f2ff9e0c62b8f1faca1a620e5d6e",
"lessThan": "886fa869153917d902784098922defa20c3a2fe5",
"versionType": "git"
},
{
"status": "affected",
"version": "7e772a93eb61cb6265bdd1c5bde17d0f2718b452",
"lessThan": "c1dd1e2b722d3f1f2e4977dad8d1be78fdfb30cb",
"versionType": "git"
},
{
"status": "affected",
"version": "7e772a93eb61cb6265bdd1c5bde17d0f2718b452",
"lessThan": "8d5fae6011260de209aaf231120e8146b14bc8e0",
"versionType": "git"
},
{
"status": "affected",
"version": "cf69b99805c263117305ac6dffbc85aaf9259d32",
"versionType": "git"
},
{
"status": "affected",
"version": "6.18.2",
"lessThan": "6.18.20",
"versionType": "semver"
},
{
"status": "affected",
"version": "6.17.13",
"lessThan": "6.18",
"versionType": "semver"
}
],
"programFiles": [
"arch/x86/events/core.c"
],
"defaultStatus": "unaffected"
},
{
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"vendor": "Linux",
"product": "Linux",
"versions": [
{
"status": "affected",
"version": "6.19"
},
{
"status": "unaffected",
"version": "0",
"lessThan": "6.19",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "6.18.20",
"versionType": "semver",
"lessThanOrEqual": "6.18.*"
},
{
"status": "unaffected",
"version": "6.19.10",
"versionType": "semver",
"lessThanOrEqual": "6.19.*"
},
{
"status": "unaffected",
"version": "7.0",
"versionType": "original_commit_for_fix",
"lessThanOrEqual": "*"
}
],
"programFiles": [
"arch/x86/events/core.c"
],
"defaultStatus": "affected"
}
]
}
],
"published": "2026-04-03T16:16:25.083",
"references": [
{
"url": "https://git.kernel.org/stable/c/886fa869153917d902784098922defa20c3a2fe5",
"tags": [
"Patch"
],
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
},
{
"url": "https://git.kernel.org/stable/c/8d5fae6011260de209aaf231120e8146b14bc8e0",
"tags": [
"Patch"
],
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
},
{
"url": "https://git.kernel.org/stable/c/c1dd1e2b722d3f1f2e4977dad8d1be78fdfb30cb",
"tags": [
"Patch"
],
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
}
],
"vulnStatus": "Analyzed",
"weaknesses": [
{
"type": "Primary",
"source": "nvd@nist.gov",
"description": [
{
"lang": "en",
"value": "CWE-476"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "In the Linux kernel, the following vulnerability has been resolved:\n\nperf/x86: Move event pointer setup earlier in x86_pmu_enable()\n\nA production AMD EPYC system crashed with a NULL pointer dereference\nin the PMU NMI handler:\n\n BUG: kernel NULL pointer dereference, address: 0000000000000198\n RIP: x86_perf_event_update+0xc/0xa0\n Call Trace:\n <NMI>\n amd_pmu_v2_handle_irq+0x1a6/0x390\n perf_event_nmi_handler+0x24/0x40\n\nThe faulting instruction is `cmpq $0x0, 0x198(%rdi)` with RDI=0,\ncorresponding to the `if (unlikely(!hwc->event_base))` check in\nx86_perf_event_update() where hwc = &event->hw and event is NULL.\n\ndrgn inspection of the vmcore on CPU 106 showed a mismatch between\ncpuc->active_mask and cpuc->events[]:\n\n active_mask: 0x1e (bits 1, 2, 3, 4)\n events[1]: 0xff1100136cbd4f38 (valid)\n events[2]: 0x0 (NULL, but active_mask bit 2 set)\n events[3]: 0xff1100076fd2cf38 (valid)\n events[4]: 0xff1100079e990a90 (valid)\n\nThe event that should occupy events[2] was found in event_list[2]\nwith hw.idx=2 and hw.state=0x0, confirming x86_pmu_start() had run\n(which clears hw.state and sets active_mask) but events[2] was\nnever populated.\n\nAnother event (event_list[0]) had hw.state=0x7 (STOPPED|UPTODATE|ARCH),\nshowing it was stopped when the PMU rescheduled events, confirming the\nthrottle-then-reschedule sequence occurred.\n\nThe root cause is commit 7e772a93eb61 (\"perf/x86: Fix NULL event access\nand potential PEBS record loss\") which moved the cpuc->events[idx]\nassignment out of x86_pmu_start() and into step 2 of x86_pmu_enable(),\nafter the PERF_HES_ARCH check. This broke any path that calls\npmu->start() without going through x86_pmu_enable() -- specifically\nthe unthrottle path:\n\n perf_adjust_freq_unthr_events()\n -> perf_event_unthrottle_group()\n -> perf_event_unthrottle()\n -> event->pmu->start(event, 0)\n -> x86_pmu_start() // sets active_mask but not events[]\n\nThe race sequence is:\n\n 1. A group of perf events overflows, triggering group throttle via\n perf_event_throttle_group(). All events are stopped: active_mask\n bits cleared, events[] preserved (x86_pmu_stop no longer clears\n events[] after commit 7e772a93eb61).\n\n 2. While still throttled (PERF_HES_STOPPED), x86_pmu_enable() runs\n due to other scheduling activity. Stopped events that need to\n move counters get PERF_HES_ARCH set and events[old_idx] cleared.\n In step 2 of x86_pmu_enable(), PERF_HES_ARCH causes these events\n to be skipped -- events[new_idx] is never set.\n\n 3. The timer tick unthrottles the group via pmu->start(). Since\n commit 7e772a93eb61 removed the events[] assignment from\n x86_pmu_start(), active_mask[new_idx] is set but events[new_idx]\n remains NULL.\n\n 4. A PMC overflow NMI fires. The handler iterates active counters,\n finds active_mask[2] set, reads events[2] which is NULL, and\n crashes dereferencing it.\n\nMove the cpuc->events[hwc->idx] assignment in x86_pmu_enable() to\nbefore the PERF_HES_ARCH check, so that events[] is populated even\nfor events that are not immediately started. This ensures the\nunthrottle path via pmu->start() always finds a valid event pointer."
},
{
"lang": "es",
"value": "En el kernel de Linux, la siguiente vulnerabilidad ha sido resuelta:\n\nperf/x86: Mover la configuración del puntero de evento antes en x86_pmu_enable()\n\nUn sistema AMD EPYC de producción falló con una desreferencia de puntero NULL en el gestor NMI de la PMU:\n\n BUG: desreferencia de puntero NULL del kernel, dirección: 0000000000000198\n RIP: x86_perf_event_update+0xc/0xa0\n Traza de Llamada:\n [NMI]\n amd_pmu_v2_handle_irq+0x1a6/0x390\n perf_event_nmi_handler+0x24/0x40\n\nLa instrucción que causó el fallo es 'cmpq $0x0, 0x198(%rdi)' con RDI=0, correspondiente a la comprobación 'if (unlikely(!hwc- event_base))' en x86_perf_event_update() donde hwc = &event- hw y event es NULL.\n\nLa inspección con drgn del vmcore en la CPU 106 mostró una discrepancia entre cpuc- active_mask y cpuc- events[]:\n\n active_mask: 0x1e (bits 1, 2, 3, 4)\n events[1]: 0xff1100136cbd4f38 (válido)\n events[2]: 0x0 (NULL, pero el bit 2 de active_mask está activado)\n events[3]: 0xff1100076fd2cf38 (válido)\n events[4]: 0xff1100079e990a90 (válido)\n\nEl evento que debería ocupar events[2] se encontró en event_list[2] con hw.idx=2 y hw.state=0x0, confirmando que x86_pmu_start() se había ejecutado (lo que borra hw.state y establece active_mask) pero events[2] nunca fue poblado.\n\nOtro evento (event_list[0]) tenía hw.state=0x7 (STOPPED|UPTODATE|ARCH), mostrando que se detuvo cuando la PMU reprogramó los eventos, confirmando que la secuencia de estrangulamiento y luego reprogramación ocurrió.\n\nLa causa raíz es el commit 7e772a93eb61 ( perf/x86: Solucionar acceso a evento NULL y posible pérdida de registro PEBS ) que movió la asignación de cpuc- events[idx] fuera de x86_pmu_start() y al paso 2 de x86_pmu_enable(), después de la comprobación PERF_HES_ARCH. Esto rompió cualquier ruta que llama a pmu- start() sin pasar por x86_pmu_enable() -- específicamente la ruta de desestrangulamiento:\n\n perf_adjust_freq_unthr_events()\n - perf_event_unthrottle_group()\n - perf_event_unthrottle()\n - event- pmu- start(event, 0)\n - x86_pmu_start() // establece active_mask pero no events[]\n\nLa secuencia de la condición de carrera es:\n\n 1. Un grupo de eventos perf desborda, lo que activa el estrangulamiento del grupo a través de perf_event_throttle_group(). Todos los eventos se detienen: los bits de active_mask se borran, events[] se preserva (x86_pmu_stop ya no borra events[] después del commit 7e772a93eb61).\n\n 2. Mientras aún está estrangulado (PERF_HES_STOPPED), x86_pmu_enable() se ejecuta debido a otra actividad de planificación. Los eventos detenidos que necesitan mover contadores obtienen PERF_HES_ARCH establecido y events[old_idx] borrado. En el paso 2 de x86_pmu_enable(), PERF_HES_ARCH hace que estos eventos se omitan -- events[new_idx] nunca se establece.\n\n 3. El tick del temporizador desestrángula el grupo a través de pmu- start(). Dado que el commit 7e772a93eb61 eliminó la asignación de events[] de x86_pmu_start(), active_mask[new_idx] se establece pero events[new_idx] permanece NULL.\n\n 4. Se dispara una NMI de desbordamiento de PMC. El gestor itera los contadores activos, encuentra active_mask[2] establecido, lee events[2] que es NULL, y falla al desreferenciarlo.\n\nMover la asignación de cpuc- events[hwc- idx] en x86_pmu_enable() a antes de la comprobación PERF_HES_ARCH, para que events[] se pueble incluso para eventos que no se inician inmediatamente. Esto asegura que la ruta de desestrangulamiento a través de pmu- start() siempre encuentre un puntero de evento válido."
}
],
"lastModified": "2026-07-21T19:10:00.107",
"configurations": [
{
"nodes": [
{
"negate": false,
"cpeMatch": [
{
"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "7C298528-1754-41BD-B4E9-84A37AB7BA32",
"versionEndExcluding": "6.18",
"versionStartIncluding": "6.17.13"
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "29896448-5F31-40A5-A909-2BC59D21590D",
"versionEndExcluding": "6.18.20",
"versionStartIncluding": "6.18.2"
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "D70DEFFE-AC47-4F3A-A2B2-2D67AB4CF3C8",
"versionEndExcluding": "6.19.10",
"versionStartIncluding": "6.19.1"
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:6.19:-:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "35C8A871-4971-433E-A046-FC9F7B7D190A"
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc1:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "F253B622-8837-4245-BCE5-A7BF8FC76A16"
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc2:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "4AE85AD8-4641-4E7C-A2F4-305E2CD9EE64"
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc3:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "F666C8D8-6538-46D4-B318-87610DE64C34"
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc4:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "02259FDA-961B-47BC-AE7F-93D7EC6E90C2"
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc5:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "58A9FEFF-C040-420D-8F0A-BFDAAA1DF258"
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc6:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "1D2315C0-D46F-4F85-9754-F9E5E11374A6"
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc7:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "512EE3A8-A590-4501-9A94-5D4B268D6138"
}
],
"operator": "OR"
}
]
}
],
"sourceIdentifier": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
}