mm/gup: reject FOLL_SPLIT_PMD with hugetlb VMAs
In the Linux kernel, the following vulnerability has been resolved: mm/gup: reject FOLL_SPLIT_PMD with hugetlb VMAs Patch series "mm: fixes for device-exclusive entries (hmm)", v2. Discussing the PageTail() call in make_device_exclusive_range() with Willy, I recently discovered [1] that device-exclusive handling does not properly work with THP, making the hmm-tests selftests fail if THPs are enabled on the system. Looking into more details, I found that hugetlb is not properly fenced, and I realized that something that was bugging me for longer -- how device-exclusive entries interact with mapcounts -- completely breaks migration/swapout/split/hwpoison handling of these folios while they have device-exclusive PTEs. The program below can be used to allocate 1 GiB worth of pages and making them device-exclusive on a kernel with CONFIG_TEST_HMM. Once they are device-exclusive, these folios cannot get swapped out (proc$pid/smaps_rollup will always indicate 1 GiB RSS no matter how much one forces memory reclaim), and when having a memory block onlined to ZONE_MOVABLE, trying to offline it will loop forever and complain about failed migration of a page that should be movable. # echo offline > /sys/devices/system/memory/memory136/state # echo online_movable > /sys/devices/system/memory/memory136/state # ./hmm-swap & ... wait until everything is device-exclusive # echo offline > /sys/devices/system/memory/memory136/state [ 285.193431][T14882] page: refcount:2 mapcount:0 mapping:00000
- Products
- Kernel
- Affected versions
- Not specified
- Fixed versions
- 6.12.23, 6.13.11, 6.14.2
- CVEs
- CVE-2025-22034
- Confirmed fixed CVEs
- Not specified
- Updated in this feed
- 5 Nov 2026, 04:24 UTC · Revision 1
Source evidence
mm/gup: reject FOLL_SPLIT_PMD with hugetlb VMAs
core-osv · Original sourcePublished 16 Apr 2025, 14:11 UTC
In the Linux kernel, the following vulnerability has been resolved: mm/gup: reject FOLL_SPLIT_PMD with hugetlb VMAs Patch series "mm: fixes for device-exclusive entries (hmm)", v2. Discussing the PageTail() call in make_device_exclusive_range() with Willy, I recently discovered [1] that device-exclusive handling does not properly work with THP, making the hmm-tests selftests fail if THPs are enabled on the system. Looking into more details, I found that hugetlb is not properly fenced, and I realized that something that was bugging me for longer -- how device-exclusive entries interact with mapcounts -- completely breaks migration/swapout/split/hwpoison handling of these folios while they have device-exclusive PTEs. The program below can be used to allocate 1 GiB worth of pages and making them device-exclusive on a kernel with CONFIG_TEST_HMM. Once they are device-exclusive, these folios cannot get swapped out (proc$pid/smaps_rollup will always indicate 1 GiB RSS no matter how much one forces memory reclaim), and when having a memory block onlined to ZONE_MOVABLE, trying to offline it will loop forever and complain about failed migration of a page that should be movable. # echo offline > /sys/devices/system/memory/memory136/state # echo online_movable > /sys/devices/system/memory/memory136/state # ./hmm-swap & ... wait until everything is device-exclusive # echo offline > /sys/devices/system/memory/memory136/state [ 285.193431][T14882] page: refcount:2 mapcount:0 mapping:00000
osv:CVE-2025-22034
Kernel
Fixed versions: 6.12.23, 6.13.11, 6.14.2
Complete source record
{
"id": "1977ad4789ca670430526831e4d486277dbd1136c344bf5ffced79d8a7116249",
"url": "https://osv.dev/vulnerability/CVE-2025-22034",
"method": "structured",
"payload": {
"kind": "advisory",
"title": "mm/gup: reject FOLL_SPLIT_PMD with hugetlb VMAs",
"summary": "In the Linux kernel, the following vulnerability has been resolved: mm/gup: reject FOLL_SPLIT_PMD with hugetlb VMAs Patch series \"mm: fixes for device-exclusive entries (hmm)\", v2. Discussing the PageTail() call in make_device_exclusive_range() with Willy, I recently discovered [1] that device-exclusive handling does not properly work with THP, making the hmm-tests selftests fail if THPs are enabled on the system. Looking into more details, I found that hugetlb is not properly fenced, and I realized that something that was bugging me for longer -- how device-exclusive entries interact with mapcounts -- completely breaks migration/swapout/split/hwpoison handling of these folios while they have device-exclusive PTEs. The program below can be used to allocate 1 GiB worth of pages and making them device-exclusive on a kernel with CONFIG_TEST_HMM. Once they are device-exclusive, these folios cannot get swapped out (proc$pid/smaps_rollup will always indicate 1 GiB RSS no matter how much one forces memory reclaim), and when having a memory block onlined to ZONE_MOVABLE, trying to offline it will loop forever and complain about failed migration of a page that should be movable. # echo offline > /sys/devices/system/memory/memory136/state # echo online_movable > /sys/devices/system/memory/memory136/state # ./hmm-swap & ... wait until everything is device-exclusive # echo offline > /sys/devices/system/memory/memory136/state [ 285.193431][T14882] page: refcount:2 mapcount:0 mapping:00000",
"evidence": "osv:CVE-2025-22034",
"products": [
{
"name": {
"value": "Kernel",
"evidence": "/affected/1/package/name"
},
"prerequisites": [],
"fixed_versions": [
{
"value": "6.12.23",
"evidence": "/affected/1/ranges"
},
{
"value": "6.13.11",
"evidence": "/affected/1/ranges"
},
{
"value": "6.14.2",
"evidence": "/affected/1/ranges"
}
],
"affected_versions": []
}
],
"severity": "high",
"structured": {
"id": "CVE-2025-22034",
"aliases": [
"CVE-2025-22034"
],
"metrics": [
{
"type": "CVSS_V3",
"score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H"
}
],
"affected": [
{
"ranges": [
{
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"type": "GIT",
"events": [
{
"introduced": "9cb28da54643ad464c47585cd5866c30b0218e67"
},
{
"fixed": "2e877ff3492267def06dd50cb165dc9ab8838e7d"
},
{
"fixed": "48d28417c66cce2f3b0ba773fcb6695a56eff220"
},
{
"fixed": "fd900832e8440046627b60697687ab5d04398008"
},
{
"fixed": "8977752c8056a6a094a279004a49722da15bace3"
}
]
}
],
"database_specific": {
"source": "https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2025-22034.json"
}
},
{
"ranges": [
{
"type": "ECOSYSTEM",
"events": [
{
"introduced": "6.10.0"
},
{
"fixed": "6.12.23"
}
]
},
{
"type": "ECOSYSTEM",
"events": [
{
"introduced": "6.13.0"
},
{
"fixed": "6.13.11"
}
]
},
{
"type": "ECOSYSTEM",
"events": [
{
"introduced": "6.14.0"
},
{
"fixed": "6.14.2"
}
]
}
],
"package": {
"name": "Kernel",
"ecosystem": "Linux"
},
"database_specific": {
"source": "https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2025-22034.json"
}
}
],
"modified": "2026-10-28T02:30:42.667756548Z",
"provider": "osv",
"references": [
"https://git.kernel.org/stable/c/2e877ff3492267def06dd50cb165dc9ab8838e7d",
"https://git.kernel.org/stable/c/48d28417c66cce2f3b0ba773fcb6695a56eff220",
"https://git.kernel.org/stable/c/8977752c8056a6a094a279004a49722da15bace3",
"https://git.kernel.org/stable/c/fd900832e8440046627b60697687ab5d04398008",
"https://github.com/CVEProject/cvelistV5/tree/main/cves/2025/22xxx/CVE-2025-22034.json",
"https://nvd.nist.gov/vuln/detail/CVE-2025-22034",
"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git"
],
"provider_metadata": {
"cna_assigner": "Linux",
"osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2025/22xxx/CVE-2025-22034.json"
}
},
"related_cves": [
{
"value": "CVE-2025-22034",
"evidence": "/id"
}
],
"security_relevant": true,
"severity_evidence": "/severity/0/score"
},
"source_id": "core-osv",
"document_id": 510238,
"published_at": "2025-04-16T14:11:53.301+00:00"
}