CVE-2026-98431 (GCVE-0-2026-98431)
Vulnerability from cvelistv5
Published
2026-10-14 13:27
Modified
2026-10-19 09:19
Severity ?
VLAI Severity ?
EPSS score ?
Summary
In the Linux kernel, the following vulnerability has been resolved:
RDMA/uverbs: Fix mmap_lock/disassociation_lock circular dependency
Commit 51976c6cd786 ("RDMA/core: Provide rdma_user_mmap_disassociate()
to disassociate mmap pages") introduced disassociation_lock to protect
new mmap registrations against uverbs_user_mmap_disassociate(), but
created an ABBA deadlock:
Thread A (mmap / fork):
mmap_lock -> disassociation_lock
Thread B (disassociate):
disassociation_lock -> mmap_lock
Fix by removing disassociation_lock entirely and using the pre-existing
hw_destroy_rwsem instead. hw_destroy_rwsem already provides the same
protection: rdma_umap_open() and ib_uverbs_mmap() both use
down_read_trylock() before registering a new VMA, so holding hw_destroy_rwsem
in uverbs_user_mmap_disassociate() is sufficient to block new registrations.
trylock is used in both mmap paths (not blocking down_read) because
mmap_lock is already held on entry, and uverbs_user_mmap_disassociate()
acquires mmap_lock internally — a blocking read would recreate the same
deadlock.
The only caller that was not taking hw_destroy_rwsem for write was
rdma_user_mmap_disassociate(). Fix it to take the rwsem per-ufile while
iterating under lists_mutex. This is safe because ib_uverbs_close()
releases hw_destroy_rwsem entirely before acquiring lists_mutex, so the
two locks are never held simultaneously.
lockdep warning:
[ 776.654252] ======================================================
[ 776.655214] WARNING: possible circular locking dependency detected
[ 776.656167] 6.18.0for-upstream_debug_94e244d9ccab #1 Not tainted
[ 776.657114] ------------------------------------------------------
[ 776.658087] devlink/14824 is trying to acquire lock:
[ 776.658879] ffff88811170c800 (&mm->mmap_lock){++++}-{4:4}, at: uverbs_user_mmap_disassociate+0x168/0x780 [ib_uverbs]
[ 776.660479]
[ 776.660479] but task is already holding lock:
[ 776.661460] ffff888142d92b08 (&file->disassociation_lock){+.+.}-{4:4}, at: uverbs_user_mmap_disassociate+0x39/0x780 [ib_uverbs]
[ 776.663177]
[ 776.663177] which lock already depends on the new lock.
[ 776.663177]
[ 776.664525]
[ 776.664525] the existing dependency chain (in reverse order) is:
[ 776.665724]
[ 776.665724] -> #2 (&file->disassociation_lock){+.+.}-{4:4}:
[ 776.666887] __mutex_lock+0x16d/0x2330
[ 776.667633] rdma_umap_open+0x129/0x280 [ib_uverbs]
[ 776.668489] dup_mmap+0xa40/0x1790
[ 776.669170] copy_process+0x5dd2/0x6170
[ 776.669933] kernel_clone+0xb6/0x610
[ 776.670636] __do_sys_clone+0xb5/0xf0
[ 776.671354] do_syscall_64+0x70/0x12e0
[ 776.672083] entry_SYSCALL_64_after_hwframe+0x4b/0x53
[ 776.672940]
[ 776.672940] -> #1 (&mm->mmap_lock/1){+.+.}-{4:4}:
[ 776.673985] down_write_nested+0x90/0x1e0
[ 776.674751] dup_mmap+0x201/0x1790
[ 776.675448] copy_process+0x5dd2/0x6170
[ 776.676180] kernel_clone+0xb6/0x610
[ 776.676904] __do_sys_clone+0xb5/0xf0
[ 776.677615] do_syscall_64+0x70/0x12e0
[ 776.678351] entry_SYSCALL_64_after_hwframe+0x4b/0x53
[ 776.679239]
[ 776.679239] -> #0 (&mm->mmap_lock){++++}-{4:4}:
[ 776.680253] __lock_acquire+0x18c6/0x2ec0
[ 776.681018] lock_acquire+0x10e/0x2e0
[ 776.681742] down_read+0x95/0x430
[ 776.682395] uverbs_user_mmap_disassociate+0x168/0x780 [ib_uverbs]
[ 776.683436] uverbs_destroy_ufile_hw+0x1ae/0x270 [ib_uverbs]
[ 776.684416] ib_uverbs_remove_one+0x22b/0x420 [ib_uverbs]
[ 776.685371] remove_client_context+0xa6/0xf0 [ib_core]
[ 776.686342] disable_device+0x12b/0x240 [ib_core]
[ 776.687249] __ib_unregister_device+0x269/0x460 [ib_core]
[ 776.688233] ib_unregister_device+0x21/0x30 [ib_core]
[ 776.689140] mlx5r_remove+0xd0/0x170 [mlx5_ib]
[ 776.689999] device_release_driver_internal+0x3b2/0x5
---truncated---
References
Impacted products
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "Linux",
"programFiles": [
"drivers/infiniband/core/rdma_core.c",
"drivers/infiniband/core/uverbs_main.c",
"include/rdma/uverbs_types.h"
],
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"vendor": "Linux",
"versions": [
{
"lessThan": "426f54f20b79c5b164a0472ad3f6ba36afcd08c6",
"status": "affected",
"version": "51976c6cd786151b6a1bdf8b8b3334beac0ba99c",
"versionType": "git"
},
{
"lessThan": "a44a3f175eaee7e5aeb6a8fed381c4a0d5f49236",
"status": "affected",
"version": "51976c6cd786151b6a1bdf8b8b3334beac0ba99c",
"versionType": "git"
},
{
"status": "affected",
"version": "0598914a950ddfa694f410e482453cdbb9265515",
"versionType": "git"
},
{
"status": "affected",
"version": "86092a4e46d161d24f8106df082e24b8efe87382",
"versionType": "git"
},
{
"lessThan": "6.12",
"status": "affected",
"version": "6.11.11",
"versionType": "semver"
},
{
"lessThan": "6.13",
"status": "affected",
"version": "6.12.2",
"versionType": "semver"
}
]
},
{
"defaultStatus": "affected",
"product": "Linux",
"programFiles": [
"drivers/infiniband/core/rdma_core.c",
"drivers/infiniband/core/uverbs_main.c",
"include/rdma/uverbs_types.h"
],
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"vendor": "Linux",
"versions": [
{
"status": "affected",
"version": "6.13"
},
{
"lessThan": "6.13",
"status": "unaffected",
"version": "0",
"versionType": "semver"
},
{
"lessThanOrEqual": "7.2.*",
"status": "unaffected",
"version": "7.2.8",
"versionType": "semver"
},
{
"lessThanOrEqual": "*",
"status": "unaffected",
"version": "7.3",
"versionType": "original_commit_for_fix"
}
]
}
],
"cpeApplicability": [
{
"nodes": [
{
"cpeMatch": [
{
"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
"versionEndExcluding": "7.2.8",
"versionStartIncluding": "6.13",
"vulnerable": true
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
"versionEndExcluding": "7.3",
"versionStartIncluding": "6.13",
"vulnerable": true
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
"versionStartIncluding": "6.11.11",
"vulnerable": true
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
"versionStartIncluding": "6.12.2",
"vulnerable": true
}
],
"negate": false,
"operator": "OR"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "In the Linux kernel, the following vulnerability has been resolved:\n\nRDMA/uverbs: Fix mmap_lock/disassociation_lock circular dependency\n\nCommit 51976c6cd786 (\"RDMA/core: Provide rdma_user_mmap_disassociate()\nto disassociate mmap pages\") introduced disassociation_lock to protect\nnew mmap registrations against uverbs_user_mmap_disassociate(), but\ncreated an ABBA deadlock:\n\n Thread A (mmap / fork):\n mmap_lock -\u003e disassociation_lock\n\n Thread B (disassociate):\n disassociation_lock -\u003e mmap_lock\n\nFix by removing disassociation_lock entirely and using the pre-existing\nhw_destroy_rwsem instead. hw_destroy_rwsem already provides the same\nprotection: rdma_umap_open() and ib_uverbs_mmap() both use\ndown_read_trylock() before registering a new VMA, so holding hw_destroy_rwsem\nin uverbs_user_mmap_disassociate() is sufficient to block new registrations.\ntrylock is used in both mmap paths (not blocking down_read) because\nmmap_lock is already held on entry, and uverbs_user_mmap_disassociate()\nacquires mmap_lock internally \u2014 a blocking read would recreate the same\ndeadlock.\n\nThe only caller that was not taking hw_destroy_rwsem for write was\nrdma_user_mmap_disassociate(). Fix it to take the rwsem per-ufile while\niterating under lists_mutex. This is safe because ib_uverbs_close()\nreleases hw_destroy_rwsem entirely before acquiring lists_mutex, so the\ntwo locks are never held simultaneously.\n\nlockdep warning:\n\n [ 776.654252] ======================================================\n [ 776.655214] WARNING: possible circular locking dependency detected\n [ 776.656167] 6.18.0for-upstream_debug_94e244d9ccab #1 Not tainted\n [ 776.657114] ------------------------------------------------------\n [ 776.658087] devlink/14824 is trying to acquire lock:\n [ 776.658879] ffff88811170c800 (\u0026mm-\u003emmap_lock){++++}-{4:4}, at: uverbs_user_mmap_disassociate+0x168/0x780 [ib_uverbs]\n [ 776.660479]\n [ 776.660479] but task is already holding lock:\n [ 776.661460] ffff888142d92b08 (\u0026file-\u003edisassociation_lock){+.+.}-{4:4}, at: uverbs_user_mmap_disassociate+0x39/0x780 [ib_uverbs]\n [ 776.663177]\n [ 776.663177] which lock already depends on the new lock.\n [ 776.663177]\n [ 776.664525]\n [ 776.664525] the existing dependency chain (in reverse order) is:\n [ 776.665724]\n [ 776.665724] -\u003e #2 (\u0026file-\u003edisassociation_lock){+.+.}-{4:4}:\n [ 776.666887] __mutex_lock+0x16d/0x2330\n [ 776.667633] rdma_umap_open+0x129/0x280 [ib_uverbs]\n [ 776.668489] dup_mmap+0xa40/0x1790\n [ 776.669170] copy_process+0x5dd2/0x6170\n [ 776.669933] kernel_clone+0xb6/0x610\n [ 776.670636] __do_sys_clone+0xb5/0xf0\n [ 776.671354] do_syscall_64+0x70/0x12e0\n [ 776.672083] entry_SYSCALL_64_after_hwframe+0x4b/0x53\n [ 776.672940]\n [ 776.672940] -\u003e #1 (\u0026mm-\u003emmap_lock/1){+.+.}-{4:4}:\n [ 776.673985] down_write_nested+0x90/0x1e0\n [ 776.674751] dup_mmap+0x201/0x1790\n [ 776.675448] copy_process+0x5dd2/0x6170\n [ 776.676180] kernel_clone+0xb6/0x610\n [ 776.676904] __do_sys_clone+0xb5/0xf0\n [ 776.677615] do_syscall_64+0x70/0x12e0\n [ 776.678351] entry_SYSCALL_64_after_hwframe+0x4b/0x53\n [ 776.679239]\n [ 776.679239] -\u003e #0 (\u0026mm-\u003emmap_lock){++++}-{4:4}:\n [ 776.680253] __lock_acquire+0x18c6/0x2ec0\n [ 776.681018] lock_acquire+0x10e/0x2e0\n [ 776.681742] down_read+0x95/0x430\n [ 776.682395] uverbs_user_mmap_disassociate+0x168/0x780 [ib_uverbs]\n [ 776.683436] uverbs_destroy_ufile_hw+0x1ae/0x270 [ib_uverbs]\n [ 776.684416] ib_uverbs_remove_one+0x22b/0x420 [ib_uverbs]\n [ 776.685371] remove_client_context+0xa6/0xf0 [ib_core]\n [ 776.686342] disable_device+0x12b/0x240 [ib_core]\n [ 776.687249] __ib_unregister_device+0x269/0x460 [ib_core]\n [ 776.688233] ib_unregister_device+0x21/0x30 [ib_core]\n [ 776.689140] mlx5r_remove+0xd0/0x170 [mlx5_ib]\n [ 776.689999] device_release_driver_internal+0x3b2/0x5\n---truncated---"
}
],
"providerMetadata": {
"dateUpdated": "2026-10-19T09:19:26.316Z",
"orgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
"shortName": "Linux"
},
"references": [
{
"url": "https://git.kernel.org/stable/c/426f54f20b79c5b164a0472ad3f6ba36afcd08c6"
},
{
"url": "https://git.kernel.org/stable/c/a44a3f175eaee7e5aeb6a8fed381c4a0d5f49236"
}
],
"title": "RDMA/uverbs: Fix mmap_lock/disassociation_lock circular dependency",
"x_generator": {
"engine": "bippy-1.2.0"
}
}
},
"cveMetadata": {
"assignerOrgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
"assignerShortName": "Linux",
"cveId": "CVE-2026-98431",
"datePublished": "2026-10-14T13:27:56.162Z",
"dateReserved": "2026-09-25T10:25:14.348Z",
"dateUpdated": "2026-10-19T09:19:26.316Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
Loading…
Loading…
Sightings
| Author | Source | Type | Date |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or seen somewhere by the user.
- Confirmed: The vulnerability is confirmed from an analyst perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: This vulnerability was exploited and seen by the user reporting the sighting.
- Patched: This vulnerability was successfully patched by the user reporting the sighting.
- Not exploited: This vulnerability was not exploited or seen by the user reporting the sighting.
- Not confirmed: The user expresses doubt about the veracity of the vulnerability.
- Not patched: This vulnerability was not successfully patched by the user reporting the sighting.
Loading…
Loading…