CVE-2025-39677 (GCVE-0-2025-39677)
Vulnerability from cvelistv5 – Published: 2025-09-05 17:20 – Updated: 2026-08-05 12:04
VLAI
EPSS
VEX
Title
net/sched: Fix backlog accounting in qdisc_dequeue_internal
Summary
In the Linux kernel, the following vulnerability has been resolved:
net/sched: Fix backlog accounting in qdisc_dequeue_internal
This issue applies for the following qdiscs: hhf, fq, fq_codel, and
fq_pie, and occurs in their change handlers when adjusting to the new
limit. The problem is the following in the values passed to the
subsequent qdisc_tree_reduce_backlog call given a tbf parent:
When the tbf parent runs out of tokens, skbs of these qdiscs will
be placed in gso_skb. Their peek handlers are qdisc_peek_dequeued,
which accounts for both qlen and backlog. However, in the case of
qdisc_dequeue_internal, ONLY qlen is accounted for when pulling
from gso_skb. This means that these qdiscs are missing a
qdisc_qstats_backlog_dec when dropping packets to satisfy the
new limit in their change handlers.
One can observe this issue with the following (with tc patched to
support a limit of 0):
export TARGET=fq
tc qdisc del dev lo root
tc qdisc add dev lo root handle 1: tbf rate 8bit burst 100b latency 1ms
tc qdisc replace dev lo handle 3: parent 1:1 $TARGET limit 1000
echo ''; echo 'add child'; tc -s -d qdisc show dev lo
ping -I lo -f -c2 -s32 -W0.001 127.0.0.1 2>&1 >/dev/null
echo ''; echo 'after ping'; tc -s -d qdisc show dev lo
tc qdisc change dev lo handle 3: parent 1:1 $TARGET limit 0
echo ''; echo 'after limit drop'; tc -s -d qdisc show dev lo
tc qdisc replace dev lo handle 2: parent 1:1 sfq
echo ''; echo 'post graft'; tc -s -d qdisc show dev lo
The second to last show command shows 0 packets but a positive
number (74) of backlog bytes. The problem becomes clearer in the
last show command, where qdisc_purge_queue triggers
qdisc_tree_reduce_backlog with the positive backlog and causes an
underflow in the tbf parent's backlog (4096 Mb instead of 0).
To fix this issue, the codepath for all clients of qdisc_dequeue_internal
has been simplified: codel, pie, hhf, fq, fq_pie, and fq_codel.
qdisc_dequeue_internal handles the backlog adjustments for all cases that
do not directly use the dequeue handler.
The old fq_codel_change limit adjustment loop accumulated the arguments to
the subsequent qdisc_tree_reduce_backlog call through the cstats field.
However, this is confusing and error prone as fq_codel_dequeue could also
potentially mutate this field (which qdisc_dequeue_internal calls in the
non gso_skb case), so we have unified the code here with other qdiscs.
Severity
7.8 (High)
Assigner
References
Impacted products
2 products
| Vendor | Product | Version | CPE status | |
|---|---|---|---|---|
| Linux | Linux |
Affected:
4b549a2ef4bef9965d97cbd992ba67930cd3e0fe , < a225f44d84b8900d679c5f5a9ea46fe9c0cc7802
(git)
Affected: 4b549a2ef4bef9965d97cbd992ba67930cd3e0fe , < 52bf272636bda69587952b35ae97690b8dc89941 (git) |
guessed | |
| Linux | Linux |
Affected:
3.5
Unaffected: 0 , < 3.5 (semver) Unaffected: 6.16.4 , ≤ 6.16.* (semver) Unaffected: 6.17 , ≤ * (original_commit_for_fix) |
guessed |
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "Linux",
"programFiles": [
"include/net/sch_generic.h",
"net/sched/sch_codel.c",
"net/sched/sch_fq.c",
"net/sched/sch_fq_codel.c",
"net/sched/sch_fq_pie.c",
"net/sched/sch_hhf.c",
"net/sched/sch_pie.c"
],
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"vendor": "Linux",
"versions": [
{
"lessThan": "a225f44d84b8900d679c5f5a9ea46fe9c0cc7802",
"status": "affected",
"version": "4b549a2ef4bef9965d97cbd992ba67930cd3e0fe",
"versionType": "git"
},
{
"lessThan": "52bf272636bda69587952b35ae97690b8dc89941",
"status": "affected",
"version": "4b549a2ef4bef9965d97cbd992ba67930cd3e0fe",
"versionType": "git"
}
]
},
{
"defaultStatus": "affected",
"product": "Linux",
"programFiles": [
"include/net/sch_generic.h",
"net/sched/sch_codel.c",
"net/sched/sch_fq.c",
"net/sched/sch_fq_codel.c",
"net/sched/sch_fq_pie.c",
"net/sched/sch_hhf.c",
"net/sched/sch_pie.c"
],
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"vendor": "Linux",
"versions": [
{
"status": "affected",
"version": "3.5"
},
{
"lessThan": "3.5",
"status": "unaffected",
"version": "0",
"versionType": "semver"
},
{
"lessThanOrEqual": "6.16.*",
"status": "unaffected",
"version": "6.16.4",
"versionType": "semver"
},
{
"lessThanOrEqual": "*",
"status": "unaffected",
"version": "6.17",
"versionType": "original_commit_for_fix"
}
]
}
],
"cpeApplicability": [
{
"nodes": [
{
"cpeMatch": [
{
"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
"versionEndExcluding": "6.16.4",
"versionStartIncluding": "3.5",
"vulnerable": true
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
"versionEndExcluding": "6.17",
"versionStartIncluding": "3.5",
"vulnerable": true
}
],
"negate": false,
"operator": "OR"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "In the Linux kernel, the following vulnerability has been resolved:\n\nnet/sched: Fix backlog accounting in qdisc_dequeue_internal\n\nThis issue applies for the following qdiscs: hhf, fq, fq_codel, and\nfq_pie, and occurs in their change handlers when adjusting to the new\nlimit. The problem is the following in the values passed to the\nsubsequent qdisc_tree_reduce_backlog call given a tbf parent:\n\n When the tbf parent runs out of tokens, skbs of these qdiscs will\n be placed in gso_skb. Their peek handlers are qdisc_peek_dequeued,\n which accounts for both qlen and backlog. However, in the case of\n qdisc_dequeue_internal, ONLY qlen is accounted for when pulling\n from gso_skb. This means that these qdiscs are missing a\n qdisc_qstats_backlog_dec when dropping packets to satisfy the\n new limit in their change handlers.\n\n One can observe this issue with the following (with tc patched to\n support a limit of 0):\n\n export TARGET=fq\n tc qdisc del dev lo root\n tc qdisc add dev lo root handle 1: tbf rate 8bit burst 100b latency 1ms\n tc qdisc replace dev lo handle 3: parent 1:1 $TARGET limit 1000\n echo \u0027\u0027; echo \u0027add child\u0027; tc -s -d qdisc show dev lo\n ping -I lo -f -c2 -s32 -W0.001 127.0.0.1 2\u003e\u00261 \u003e/dev/null\n echo \u0027\u0027; echo \u0027after ping\u0027; tc -s -d qdisc show dev lo\n tc qdisc change dev lo handle 3: parent 1:1 $TARGET limit 0\n echo \u0027\u0027; echo \u0027after limit drop\u0027; tc -s -d qdisc show dev lo\n tc qdisc replace dev lo handle 2: parent 1:1 sfq\n echo \u0027\u0027; echo \u0027post graft\u0027; tc -s -d qdisc show dev lo\n\n The second to last show command shows 0 packets but a positive\n number (74) of backlog bytes. The problem becomes clearer in the\n last show command, where qdisc_purge_queue triggers\n qdisc_tree_reduce_backlog with the positive backlog and causes an\n underflow in the tbf parent\u0027s backlog (4096 Mb instead of 0).\n\nTo fix this issue, the codepath for all clients of qdisc_dequeue_internal\nhas been simplified: codel, pie, hhf, fq, fq_pie, and fq_codel.\nqdisc_dequeue_internal handles the backlog adjustments for all cases that\ndo not directly use the dequeue handler.\n\nThe old fq_codel_change limit adjustment loop accumulated the arguments to\nthe subsequent qdisc_tree_reduce_backlog call through the cstats field.\nHowever, this is confusing and error prone as fq_codel_dequeue could also\npotentially mutate this field (which qdisc_dequeue_internal calls in the\nnon gso_skb case), so we have unified the code here with other qdiscs."
}
],
"metrics": [
{
"cvssV3_1": {
"baseScore": 7.8,
"baseSeverity": "HIGH",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"scenarios": [
{
"lang": "en",
"value": "AV:L - The vulnerability is reached exclusively through RTM_NEWQDISC/tc netlink messages that invoke the qdisc -\u003echange() handlers on a local netlink socket. Per kernel scoring guidance, tc/netlink qdisc configuration is a local attack vector.\nAC:L - The commit message contains a fully deterministic reproducer (tbf parent + fq child, ping to fill the queue, then change the limit and graft) with no race, no memory-layout dependence, and no non-default kernel config \u2014 fq and fq_codel are built in on virtually all distributions and fq_codel is the default qdisc.\nPR:L - tc_modify_qdisc is gated only by netlink_net_capable(skb, CAP_NET_ADMIN), which is evaluated against net-\u003euser_ns, so an unprivileged user obtains it with `unshare -Urn` and configures qdiscs on the loopback device in the new namespace. No real root in the init namespace is needed.\nUI:N - The attacker performs the entire sequence \u2014 creating the qdisc hierarchy, generating traffic into the queue, changing the limit, and grafting \u2014 with no victim action required.\nS:U - The corrupted state and its consequences are confined to the kernel\u0027s qdisc tree within the same security authority; there is no VM, IOMMU, or sandbox boundary crossed.\nC:H - The qdisc queue-accounting desync is the same primitive that has repeatedly escalated to use-after-free on qdisc class objects via wrongly-driven cops-\u003eqlen_notify() (CVE-2025-21703, CVE-2025-38000/38001), and a UAF on those objects yields attacker-controlled reads of freed kernel memory.\nI:H - The bug produces an attacker-controlled u32 underflow of qstats.backlog that propagates through qdisc_tree_reduce_backlog into every ancestor qdisc, corrupting kernel state that drives class active-list maintenance and enqueue/drop decisions; the resulting accounting desync is exploitable for heap corruption in the same manner as the prior net/sched accounting CVEs.\nA:H - The underflowed backlog permanently blackholes the device (bfifo/plug limit checks and sch_red\u0027s qavg see a ~4 GiB queue and drop everything), and the unguarded pre-fix trim loops give an outright kernel oops via qdisc_pkt_len(NULL) in codel/pie/fq_pie/fq_codel or an infinite loop in hhf while holding sch_tree_lock and RTNL."
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-05T12:04:29.256Z",
"orgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
"shortName": "Linux"
},
"references": [
{
"url": "https://git.kernel.org/stable/c/a225f44d84b8900d679c5f5a9ea46fe9c0cc7802"
},
{
"url": "https://git.kernel.org/stable/c/52bf272636bda69587952b35ae97690b8dc89941"
}
],
"title": "net/sched: Fix backlog accounting in qdisc_dequeue_internal",
"x_generator": {
"engine": "bippy-1.2.0"
}
}
},
"cveMetadata": {
"assignerOrgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
"assignerShortName": "Linux",
"cveId": "CVE-2025-39677",
"datePublished": "2025-09-05T17:20:43.145Z",
"dateReserved": "2025-04-16T07:20:57.112Z",
"dateUpdated": "2026-08-05T12:04:29.256Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2",
"vulnerability-lookup:meta": {
"epss": {
"cve": "CVE-2025-39677",
"date": "2026-09-20",
"epss": "0.00144",
"percentile": "0.04085"
},
"microsoft_vex": {
"current_release_date": "2026-08-10T14:56:14.000Z",
"cve": "CVE-2025-39677",
"id": "msrc_CVE-2025-39677",
"initial_release_date": "2025-09-07T01:09:12.000Z",
"product_status:known_affected": "23",
"source": "Microsoft CSAF VEX",
"status": "final",
"title": "net/sched: Fix backlog accounting in qdisc_dequeue_internal",
"url": "https://msrc.microsoft.com/csaf/vex/2025/msrc_cve-2025-39677.json",
"version": "19"
},
"nvd": {
"cve": {
"affected": [
{
"affectedData": [
{
"defaultStatus": "unaffected",
"product": "Linux",
"programFiles": [
"include/net/sch_generic.h",
"net/sched/sch_codel.c",
"net/sched/sch_fq.c",
"net/sched/sch_fq_codel.c",
"net/sched/sch_fq_pie.c",
"net/sched/sch_hhf.c",
"net/sched/sch_pie.c"
],
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"vendor": "Linux",
"versions": [
{
"lessThan": "a225f44d84b8900d679c5f5a9ea46fe9c0cc7802",
"status": "affected",
"version": "4b549a2ef4bef9965d97cbd992ba67930cd3e0fe",
"versionType": "git"
},
{
"lessThan": "52bf272636bda69587952b35ae97690b8dc89941",
"status": "affected",
"version": "4b549a2ef4bef9965d97cbd992ba67930cd3e0fe",
"versionType": "git"
}
]
},
{
"defaultStatus": "affected",
"product": "Linux",
"programFiles": [
"include/net/sch_generic.h",
"net/sched/sch_codel.c",
"net/sched/sch_fq.c",
"net/sched/sch_fq_codel.c",
"net/sched/sch_fq_pie.c",
"net/sched/sch_hhf.c",
"net/sched/sch_pie.c"
],
"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
"vendor": "Linux",
"versions": [
{
"status": "affected",
"version": "3.5"
},
{
"lessThan": "3.5",
"status": "unaffected",
"version": "0",
"versionType": "semver"
},
{
"lessThanOrEqual": "6.16.*",
"status": "unaffected",
"version": "6.16.4",
"versionType": "semver"
},
{
"lessThanOrEqual": "*",
"status": "unaffected",
"version": "6.17",
"versionType": "original_commit_for_fix"
}
]
}
],
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"
}
],
"configurations": [
{
"nodes": [
{
"cpeMatch": [
{
"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
"matchCriteriaId": "6C2046DB-71F9-451C-B88C-B00FB2E419A2",
"versionEndExcluding": "6.16.4",
"versionStartIncluding": "3.5",
"vulnerable": true
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:6.17:rc1:*:*:*:*:*:*",
"matchCriteriaId": "327D22EF-390B-454C-BD31-2ED23C998A1C",
"vulnerable": true
},
{
"criteria": "cpe:2.3:o:linux:linux_kernel:6.17:rc2:*:*:*:*:*:*",
"matchCriteriaId": "C730CD9A-D969-4A8E-9522-162AAF7C0EE9",
"vulnerable": true
}
],
"negate": false,
"operator": "OR"
}
]
}
],
"cveTags": [],
"descriptions": [
{
"lang": "en",
"value": "In the Linux kernel, the following vulnerability has been resolved:\n\nnet/sched: Fix backlog accounting in qdisc_dequeue_internal\n\nThis issue applies for the following qdiscs: hhf, fq, fq_codel, and\nfq_pie, and occurs in their change handlers when adjusting to the new\nlimit. The problem is the following in the values passed to the\nsubsequent qdisc_tree_reduce_backlog call given a tbf parent:\n\n When the tbf parent runs out of tokens, skbs of these qdiscs will\n be placed in gso_skb. Their peek handlers are qdisc_peek_dequeued,\n which accounts for both qlen and backlog. However, in the case of\n qdisc_dequeue_internal, ONLY qlen is accounted for when pulling\n from gso_skb. This means that these qdiscs are missing a\n qdisc_qstats_backlog_dec when dropping packets to satisfy the\n new limit in their change handlers.\n\n One can observe this issue with the following (with tc patched to\n support a limit of 0):\n\n export TARGET=fq\n tc qdisc del dev lo root\n tc qdisc add dev lo root handle 1: tbf rate 8bit burst 100b latency 1ms\n tc qdisc replace dev lo handle 3: parent 1:1 $TARGET limit 1000\n echo \u0027\u0027; echo \u0027add child\u0027; tc -s -d qdisc show dev lo\n ping -I lo -f -c2 -s32 -W0.001 127.0.0.1 2\u003e\u00261 \u003e/dev/null\n echo \u0027\u0027; echo \u0027after ping\u0027; tc -s -d qdisc show dev lo\n tc qdisc change dev lo handle 3: parent 1:1 $TARGET limit 0\n echo \u0027\u0027; echo \u0027after limit drop\u0027; tc -s -d qdisc show dev lo\n tc qdisc replace dev lo handle 2: parent 1:1 sfq\n echo \u0027\u0027; echo \u0027post graft\u0027; tc -s -d qdisc show dev lo\n\n The second to last show command shows 0 packets but a positive\n number (74) of backlog bytes. The problem becomes clearer in the\n last show command, where qdisc_purge_queue triggers\n qdisc_tree_reduce_backlog with the positive backlog and causes an\n underflow in the tbf parent\u0027s backlog (4096 Mb instead of 0).\n\nTo fix this issue, the codepath for all clients of qdisc_dequeue_internal\nhas been simplified: codel, pie, hhf, fq, fq_pie, and fq_codel.\nqdisc_dequeue_internal handles the backlog adjustments for all cases that\ndo not directly use the dequeue handler.\n\nThe old fq_codel_change limit adjustment loop accumulated the arguments to\nthe subsequent qdisc_tree_reduce_backlog call through the cstats field.\nHowever, this is confusing and error prone as fq_codel_dequeue could also\npotentially mutate this field (which qdisc_dequeue_internal calls in the\nnon gso_skb case), so we have unified the code here with other qdiscs."
}
],
"id": "CVE-2025-39677",
"lastModified": "2026-07-30T06:23:39.233",
"metrics": {
"cvssMetricV31": [
{
"cvssData": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "LOW",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"exploitabilityScore": 1.8,
"impactScore": 5.9,
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
"type": "Secondary"
},
{
"cvssData": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 5.5,
"baseSeverity": "MEDIUM",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "LOW",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
"version": "3.1"
},
"exploitabilityScore": 1.8,
"impactScore": 3.6,
"source": "nvd@nist.gov",
"type": "Primary"
}
]
},
"published": "2025-09-05T18:15:44.043",
"references": [
{
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
"tags": [
"Patch"
],
"url": "https://git.kernel.org/stable/c/52bf272636bda69587952b35ae97690b8dc89941"
},
{
"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
"tags": [
"Patch"
],
"url": "https://git.kernel.org/stable/c/a225f44d84b8900d679c5f5a9ea46fe9c0cc7802"
}
],
"sourceIdentifier": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
"vulnStatus": "Modified",
"weaknesses": [
{
"description": [
{
"lang": "en",
"value": "NVD-CWE-noinfo"
}
],
"source": "nvd@nist.gov",
"type": "Primary"
}
]
}
},
"redhat_vex": {
"aggregate_severity": "Moderate",
"current_release_date": "2026-07-30T08:59:15+00:00",
"cve": "CVE-2025-39677",
"id": "CVE-2025-39677",
"initial_release_date": "2025-09-05T00:00:00+00:00",
"product_status:known_affected": "260",
"product_status:known_not_affected": "14",
"source": "Red Hat CSAF VEX",
"status": "final",
"title": "kernel: net/sched: Fix backlog accounting in qdisc_dequeue_internal",
"url": "https://security.access.redhat.com/data/csaf/v2/vex/2025/cve-2025-39677.json",
"version": "3"
}
}
}
Loading…
Loading…
Experimental. This forecast is provided for visualization only and may change without notice. Do not use it for operational decisions.
Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Loading…
The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the
vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base
model, served locally by ML-Gateway.
They have not been verified by an analyst and are provided for guidance only.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Loading…
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.
Loading…