CWE-476
AllowedNULL Pointer Dereference
Abstraction: Base · Status: Stable
The product dereferences a pointer that it expects to be valid but is NULL.
6948 vulnerabilities reference this CWE, most recent first.
GHSA-VQPV-2JR8-2HMP
Vulnerability from github – Published: 2026-05-08 15:31 – Updated: 2026-05-15 15:30In the Linux kernel, the following vulnerability has been resolved:
drm/panel: Fix a possible null-pointer dereference in jdi_panel_dsi_remove()
In jdi_panel_dsi_remove(), jdi is explicitly checked, indicating that it may be NULL:
if (!jdi) mipi_dsi_detach(dsi);
However, when jdi is NULL, the function does not return and continues by calling jdi_panel_disable():
err = jdi_panel_disable(&jdi->base);
Inside jdi_panel_disable(), jdi is dereferenced unconditionally, which can lead to a NULL-pointer dereference:
struct jdi_panel *jdi = to_panel_jdi(panel); backlight_disable(jdi->backlight);
To prevent such a potential NULL-pointer dereference, return early from jdi_panel_dsi_remove() when jdi is NULL.
{
"affected": [],
"aliases": [
"CVE-2026-43300"
],
"database_specific": {
"cwe_ids": [
"CWE-476"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-05-08T14:16:37Z",
"severity": "MODERATE"
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/panel: Fix a possible null-pointer dereference in jdi_panel_dsi_remove()\n\nIn jdi_panel_dsi_remove(), jdi is explicitly checked, indicating that it\nmay be NULL:\n\n if (!jdi)\n mipi_dsi_detach(dsi);\n\nHowever, when jdi is NULL, the function does not return and continues by\ncalling jdi_panel_disable():\n\n err = jdi_panel_disable(\u0026jdi-\u003ebase);\n\nInside jdi_panel_disable(), jdi is dereferenced unconditionally, which can\nlead to a NULL-pointer dereference:\n\n struct jdi_panel *jdi = to_panel_jdi(panel);\n backlight_disable(jdi-\u003ebacklight);\n\nTo prevent such a potential NULL-pointer dereference, return early from\njdi_panel_dsi_remove() when jdi is NULL.",
"id": "GHSA-vqpv-2jr8-2hmp",
"modified": "2026-05-15T15:30:33Z",
"published": "2026-05-08T15:31:22Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-43300"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/2f5427d8726b22b807beec248d7d6bf88e291e0b"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/83ce0085fabf757b039322928188ad78e962d609"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/95eed73b871111123a8b1d31cb1fce7e902e49ea"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/ec2f37bbb733cdd7ed7d04171fca728a532414d5"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
]
}
GHSA-VQQW-285R-PW6X
Vulnerability from github – Published: 2026-04-03 18:31 – Updated: 2026-04-23 21:31In the Linux kernel, the following vulnerability has been resolved:
arm_mpam: Fix null pointer dereference when restoring bandwidth counters
When an MSC supporting memory bandwidth monitoring is brought offline and then online, mpam_restore_mbwu_state() calls __ris_msmon_read() via ipi to restore the configuration of the bandwidth counters. It doesn't care about the value read, mbwu_arg.val, and doesn't set it leading to a null pointer dereference when __ris_msmon_read() adds to it. This results in a kernel oops with a call trace such as:
Call trace: __ris_msmon_read+0x19c/0x64c (P) mpam_restore_mbwu_state+0xa0/0xe8 smp_call_on_cpu_callback+0x1c/0x38 process_one_work+0x154/0x4b4 worker_thread+0x188/0x310 kthread+0x11c/0x130 ret_from_fork+0x10/0x20
Provide a local variable for val to avoid __ris_msmon_read() dereferencing a null pointer when adding to val.
{
"affected": [],
"aliases": [
"CVE-2026-23433"
],
"database_specific": {
"cwe_ids": [
"CWE-476"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-04-03T16:16:24Z",
"severity": "MODERATE"
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\narm_mpam: Fix null pointer dereference when restoring bandwidth counters\n\nWhen an MSC supporting memory bandwidth monitoring is brought offline and\nthen online, mpam_restore_mbwu_state() calls __ris_msmon_read() via ipi to\nrestore the configuration of the bandwidth counters. It doesn\u0027t care about\nthe value read, mbwu_arg.val, and doesn\u0027t set it leading to a null pointer\ndereference when __ris_msmon_read() adds to it. This results in a kernel\noops with a call trace such as:\n\nCall trace:\n__ris_msmon_read+0x19c/0x64c (P)\nmpam_restore_mbwu_state+0xa0/0xe8\nsmp_call_on_cpu_callback+0x1c/0x38\nprocess_one_work+0x154/0x4b4\nworker_thread+0x188/0x310\nkthread+0x11c/0x130\nret_from_fork+0x10/0x20\n\nProvide a local variable for val to avoid __ris_msmon_read() dereferencing\na null pointer when adding to val.",
"id": "GHSA-vqqw-285r-pw6x",
"modified": "2026-04-23T21:31:18Z",
"published": "2026-04-03T18:31:21Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-23433"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/4ad79c874e53ebb7fe3b8ae7ac6c858a2121f415"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/ac3e12bc195786d3d44d730b5b2259fd36191848"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
]
}
GHSA-VQXQ-CHHR-677M
Vulnerability from github – Published: 2025-07-07 03:30 – Updated: 2025-07-07 03:30Null pointer dereference vulnerability in the PDF preview module Impact: Successful exploitation of this vulnerability may affect function stability.
{
"affected": [],
"aliases": [
"CVE-2025-53182"
],
"database_specific": {
"cwe_ids": [
"CWE-122",
"CWE-476"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2025-07-07T03:15:29Z",
"severity": "MODERATE"
},
"details": "Null pointer dereference vulnerability in the PDF preview module\nImpact: Successful exploitation of this vulnerability may affect function stability.",
"id": "GHSA-vqxq-chhr-677m",
"modified": "2025-07-07T03:30:23Z",
"published": "2025-07-07T03:30:23Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2025-53182"
},
{
"type": "WEB",
"url": "https://consumer.huawei.com/en/support/bulletin/2025/7"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
]
}
GHSA-VR29-W5CG-Q4XV
Vulnerability from github – Published: 2024-12-07 00:31 – Updated: 2024-12-12 03:33Open Robotics Robotic Operating System 2 ROS2 navigation2 v.humble was discovered to contain a NULL pointer dereference via the component computeControl().
{
"affected": [],
"aliases": [
"CVE-2024-44853"
],
"database_specific": {
"cwe_ids": [
"CWE-476"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2024-12-06T22:15:21Z",
"severity": "HIGH"
},
"details": "Open Robotics Robotic Operating System 2 ROS2 navigation2 v.humble was discovered to contain a NULL pointer dereference via the component computeControl().",
"id": "GHSA-vr29-w5cg-q4xv",
"modified": "2024-12-12T03:33:00Z",
"published": "2024-12-07T00:31:04Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2024-44853"
},
{
"type": "WEB",
"url": "https://github.com/ros-navigation/navigation2/issues/4547"
},
{
"type": "WEB",
"url": "https://github.com/ros-navigation/navigation2/pull/4548"
},
{
"type": "WEB",
"url": "https://github.com/GoesM/ROS-CVE-CNVDs"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
]
}
GHSA-VR35-F6M3-RH89
Vulnerability from github – Published: 2024-05-21 18:31 – Updated: 2025-01-06 21:30In the Linux kernel, the following vulnerability has been resolved:
tls: fix NULL deref on tls_sw_splice_eof() with empty record
syzkaller discovered that if tls_sw_splice_eof() is executed as part of
sendfile() when the plaintext/ciphertext sk_msg are empty, the send path
gets confused because the empty ciphertext buffer does not have enough
space for the encryption overhead. This causes tls_push_record() to go on
the split = true path (which is only supposed to be used when interacting
with an attached BPF program), and then get further confused and hit the
tls_merge_open_record() path, which then assumes that there must be at
least one populated buffer element, leading to a NULL deref.
It is possible to have empty plaintext/ciphertext buffers if we previously bailed from tls_sw_sendmsg_locked() via the tls_trim_both_msgs() path. tls_sw_push_pending_record() already handles this case correctly; let's do the same check in tls_sw_splice_eof().
{
"affected": [],
"aliases": [
"CVE-2023-52767"
],
"database_specific": {
"cwe_ids": [
"CWE-476"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2024-05-21T16:15:15Z",
"severity": "MODERATE"
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\ntls: fix NULL deref on tls_sw_splice_eof() with empty record\n\nsyzkaller discovered that if tls_sw_splice_eof() is executed as part of\nsendfile() when the plaintext/ciphertext sk_msg are empty, the send path\ngets confused because the empty ciphertext buffer does not have enough\nspace for the encryption overhead. This causes tls_push_record() to go on\nthe `split = true` path (which is only supposed to be used when interacting\nwith an attached BPF program), and then get further confused and hit the\ntls_merge_open_record() path, which then assumes that there must be at\nleast one populated buffer element, leading to a NULL deref.\n\nIt is possible to have empty plaintext/ciphertext buffers if we previously\nbailed from tls_sw_sendmsg_locked() via the tls_trim_both_msgs() path.\ntls_sw_push_pending_record() already handles this case correctly; let\u0027s do\nthe same check in tls_sw_splice_eof().",
"id": "GHSA-vr35-f6m3-rh89",
"modified": "2025-01-06T21:30:49Z",
"published": "2024-05-21T18:31:20Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2023-52767"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/2214e2bb5489145aba944874d0ee1652a0a63dc8"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/53f2cb491b500897a619ff6abd72f565933760f0"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/944900fe2736c07288efe2d9394db4d3ca23f2c9"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
]
}
GHSA-VR4R-JM4F-GRC9
Vulnerability from github – Published: 2022-05-24 17:22 – Updated: 2023-01-27 21:31Appweb before 7.2.2 and 8.x before 8.1.0, when built with CGI support, mishandles an HTTP request with a Range header that lacks an exact range. This may result in a NULL pointer dereference and cause a denial of service.
{
"affected": [],
"aliases": [
"CVE-2020-15689"
],
"database_specific": {
"cwe_ids": [
"CWE-476"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2020-07-13T14:15:00Z",
"severity": "MODERATE"
},
"details": "Appweb before 7.2.2 and 8.x before 8.1.0, when built with CGI support, mishandles an HTTP request with a Range header that lacks an exact range. This may result in a NULL pointer dereference and cause a denial of service.",
"id": "GHSA-vr4r-jm4f-grc9",
"modified": "2023-01-27T21:31:11Z",
"published": "2022-05-24T17:22:49Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2020-15689"
},
{
"type": "WEB",
"url": "https://github.com/embedthis/appweb-gpl/issues/2"
},
{
"type": "WEB",
"url": "https://github.com/embedthis/appweb-gpl/issues/4"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
]
}
GHSA-VR5F-5953-GFJ5
Vulnerability from github – Published: 2024-04-28 15:30 – Updated: 2025-09-18 15:30In the Linux kernel, the following vulnerability has been resolved:
net/sched: taprio: avoid disabling offload when it was never enabled
In an incredibly strange API design decision, qdisc->destroy() gets called even if qdisc->init() never succeeded, not exclusively since commit 87b60cfacf9f ("net_sched: fix error recovery at qdisc creation"), but apparently also earlier (in the case of qdisc_create_dflt()).
The taprio qdisc does not fully acknowledge this when it attempts full offload, because it starts off with q->flags = TAPRIO_FLAGS_INVALID in taprio_init(), then it replaces q->flags with TCA_TAPRIO_ATTR_FLAGS parsed from netlink (in taprio_change(), tail called from taprio_init()).
But in taprio_destroy(), we call taprio_disable_offload(), and this determines what to do based on FULL_OFFLOAD_IS_ENABLED(q->flags).
But looking at the implementation of FULL_OFFLOAD_IS_ENABLED() (a bitwise check of bit 1 in q->flags), it is invalid to call this macro on q->flags when it contains TAPRIO_FLAGS_INVALID, because that is set to U32_MAX, and therefore FULL_OFFLOAD_IS_ENABLED() will return true on an invalid set of flags.
As a result, it is possible to crash the kernel if user space forces an error between setting q->flags = TAPRIO_FLAGS_INVALID, and the calling of taprio_enable_offload(). This is because drivers do not expect the offload to be disabled when it was never enabled.
The error that we force here is to attach taprio as a non-root qdisc, but instead as child of an mqprio root qdisc:
$ tc qdisc add dev swp0 root handle 1: \ mqprio num_tc 8 map 0 1 2 3 4 5 6 7 \ queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 hw 0 $ tc qdisc replace dev swp0 parent 1:1 \ taprio num_tc 8 map 0 1 2 3 4 5 6 7 \ queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 base-time 0 \ sched-entry S 0x7f 990000 sched-entry S 0x80 100000 \ flags 0x0 clockid CLOCK_TAI Unable to handle kernel paging request at virtual address fffffffffffffff8 [fffffffffffffff8] pgd=0000000000000000, p4d=0000000000000000 Internal error: Oops: 96000004 [#1] PREEMPT SMP Call trace: taprio_dump+0x27c/0x310 vsc9959_port_setup_tc+0x1f4/0x460 felix_port_setup_tc+0x24/0x3c dsa_slave_setup_tc+0x54/0x27c taprio_disable_offload.isra.0+0x58/0xe0 taprio_destroy+0x80/0x104 qdisc_create+0x240/0x470 tc_modify_qdisc+0x1fc/0x6b0 rtnetlink_rcv_msg+0x12c/0x390 netlink_rcv_skb+0x5c/0x130 rtnetlink_rcv+0x1c/0x2c
Fix this by keeping track of the operations we made, and undo the offload only if we actually did it.
I've added "bool offloaded" inside a 4 byte hole between "int clockid" and "atomic64_t picos_per_byte". Now the first cache line looks like below:
$ pahole -C taprio_sched net/sched/sch_taprio.o struct taprio_sched { struct Qdisc * * qdiscs; /* 0 8 / struct Qdisc * root; / 8 8 / u32 flags; / 16 4 / enum tk_offsets tk_offset; / 20 4 / int clockid; / 24 4 / bool offloaded; / 28 1 */
/* XXX 3 bytes hole, try to pack */
atomic64_t picos_per_byte; /* 32 0 */
/* XXX 8 bytes hole, try to pack */
spinlock_t current_entry_lock; /* 40 0 */
/* XXX 8 bytes hole, try to pack */
struct sched_entry * current_entry; /* 48 8 */
struct sched_gate_list * oper_sched; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
{
"affected": [],
"aliases": [
"CVE-2022-48644"
],
"database_specific": {
"cwe_ids": [
"CWE-476"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2024-04-28T13:15:07Z",
"severity": "MODERATE"
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\nnet/sched: taprio: avoid disabling offload when it was never enabled\n\nIn an incredibly strange API design decision, qdisc-\u003edestroy() gets\ncalled even if qdisc-\u003einit() never succeeded, not exclusively since\ncommit 87b60cfacf9f (\"net_sched: fix error recovery at qdisc creation\"),\nbut apparently also earlier (in the case of qdisc_create_dflt()).\n\nThe taprio qdisc does not fully acknowledge this when it attempts full\noffload, because it starts off with q-\u003eflags = TAPRIO_FLAGS_INVALID in\ntaprio_init(), then it replaces q-\u003eflags with TCA_TAPRIO_ATTR_FLAGS\nparsed from netlink (in taprio_change(), tail called from taprio_init()).\n\nBut in taprio_destroy(), we call taprio_disable_offload(), and this\ndetermines what to do based on FULL_OFFLOAD_IS_ENABLED(q-\u003eflags).\n\nBut looking at the implementation of FULL_OFFLOAD_IS_ENABLED()\n(a bitwise check of bit 1 in q-\u003eflags), it is invalid to call this macro\non q-\u003eflags when it contains TAPRIO_FLAGS_INVALID, because that is set\nto U32_MAX, and therefore FULL_OFFLOAD_IS_ENABLED() will return true on\nan invalid set of flags.\n\nAs a result, it is possible to crash the kernel if user space forces an\nerror between setting q-\u003eflags = TAPRIO_FLAGS_INVALID, and the calling\nof taprio_enable_offload(). This is because drivers do not expect the\noffload to be disabled when it was never enabled.\n\nThe error that we force here is to attach taprio as a non-root qdisc,\nbut instead as child of an mqprio root qdisc:\n\n$ tc qdisc add dev swp0 root handle 1: \\\n\tmqprio num_tc 8 map 0 1 2 3 4 5 6 7 \\\n\tqueues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 hw 0\n$ tc qdisc replace dev swp0 parent 1:1 \\\n\ttaprio num_tc 8 map 0 1 2 3 4 5 6 7 \\\n\tqueues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 base-time 0 \\\n\tsched-entry S 0x7f 990000 sched-entry S 0x80 100000 \\\n\tflags 0x0 clockid CLOCK_TAI\nUnable to handle kernel paging request at virtual address fffffffffffffff8\n[fffffffffffffff8] pgd=0000000000000000, p4d=0000000000000000\nInternal error: Oops: 96000004 [#1] PREEMPT SMP\nCall trace:\n taprio_dump+0x27c/0x310\n vsc9959_port_setup_tc+0x1f4/0x460\n felix_port_setup_tc+0x24/0x3c\n dsa_slave_setup_tc+0x54/0x27c\n taprio_disable_offload.isra.0+0x58/0xe0\n taprio_destroy+0x80/0x104\n qdisc_create+0x240/0x470\n tc_modify_qdisc+0x1fc/0x6b0\n rtnetlink_rcv_msg+0x12c/0x390\n netlink_rcv_skb+0x5c/0x130\n rtnetlink_rcv+0x1c/0x2c\n\nFix this by keeping track of the operations we made, and undo the\noffload only if we actually did it.\n\nI\u0027ve added \"bool offloaded\" inside a 4 byte hole between \"int clockid\"\nand \"atomic64_t picos_per_byte\". Now the first cache line looks like\nbelow:\n\n$ pahole -C taprio_sched net/sched/sch_taprio.o\nstruct taprio_sched {\n struct Qdisc * * qdiscs; /* 0 8 */\n struct Qdisc * root; /* 8 8 */\n u32 flags; /* 16 4 */\n enum tk_offsets tk_offset; /* 20 4 */\n int clockid; /* 24 4 */\n bool offloaded; /* 28 1 */\n\n /* XXX 3 bytes hole, try to pack */\n\n atomic64_t picos_per_byte; /* 32 0 */\n\n /* XXX 8 bytes hole, try to pack */\n\n spinlock_t current_entry_lock; /* 40 0 */\n\n /* XXX 8 bytes hole, try to pack */\n\n struct sched_entry * current_entry; /* 48 8 */\n struct sched_gate_list * oper_sched; /* 56 8 */\n /* --- cacheline 1 boundary (64 bytes) --- */",
"id": "GHSA-vr5f-5953-gfj5",
"modified": "2025-09-18T15:30:21Z",
"published": "2024-04-28T15:30:29Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2022-48644"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/586def6ebed195f3594a4884f7c5334d0e1ad1bb"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/c7c9c7eb305ab8b4e93e4e4e1b78d8cfcbc26323"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/d12a1eb07003e597077329767c6aa86a7e972c76"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/db46e3a88a09c5cf7e505664d01da7238cd56c92"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/f58e43184226e5e9662088ccf1389e424a3a4cbd"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
]
}
GHSA-VR5F-V75P-G4QW
Vulnerability from github – Published: 2024-03-25 09:32 – Updated: 2024-11-07 18:31In the Linux kernel, the following vulnerability has been resolved:
iommu/amd: Clear DMA ops when switching domain
Since commit 08a27c1c3ecf ("iommu: Add support to change default domain of an iommu group") a user can switch a device between IOMMU and direct DMA through sysfs. This doesn't work for AMD IOMMU at the moment because dev->dma_ops is not cleared when switching from a DMA to an identity IOMMU domain. The DMA layer thus attempts to use the dma-iommu ops on an identity domain, causing an oops:
# echo 0000:00:05.0 > /sys/sys/bus/pci/drivers/e1000e/unbind # echo identity > /sys/bus/pci/devices/0000:00:05.0/iommu_group/type # echo 0000:00:05.0 > /sys/sys/bus/pci/drivers/e1000e/bind ... BUG: kernel NULL pointer dereference, address: 0000000000000028 ... Call Trace: iommu_dma_alloc e1000e_setup_tx_resources e1000e_open
Since iommu_change_dev_def_domain() calls probe_finalize() again, clear the dma_ops there like Vt-d does.
{
"affected": [],
"aliases": [
"CVE-2021-47140"
],
"database_specific": {
"cwe_ids": [
"CWE-476"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2024-03-25T09:15:08Z",
"severity": "MODERATE"
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\niommu/amd: Clear DMA ops when switching domain\n\nSince commit 08a27c1c3ecf (\"iommu: Add support to change default domain\nof an iommu group\") a user can switch a device between IOMMU and direct\nDMA through sysfs. This doesn\u0027t work for AMD IOMMU at the moment because\ndev-\u003edma_ops is not cleared when switching from a DMA to an identity\nIOMMU domain. The DMA layer thus attempts to use the dma-iommu ops on an\nidentity domain, causing an oops:\n\n # echo 0000:00:05.0 \u003e /sys/sys/bus/pci/drivers/e1000e/unbind\n # echo identity \u003e /sys/bus/pci/devices/0000:00:05.0/iommu_group/type\n # echo 0000:00:05.0 \u003e /sys/sys/bus/pci/drivers/e1000e/bind\n ...\n BUG: kernel NULL pointer dereference, address: 0000000000000028\n ...\n Call Trace:\n iommu_dma_alloc\n e1000e_setup_tx_resources\n e1000e_open\n\nSince iommu_change_dev_def_domain() calls probe_finalize() again, clear\nthe dma_ops there like Vt-d does.",
"id": "GHSA-vr5f-v75p-g4qw",
"modified": "2024-11-07T18:31:20Z",
"published": "2024-03-25T09:32:35Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2021-47140"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/d6177a6556f853785867e2ec6d5b7f4906f0d809"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/f3f2cf46291a693eab21adb94171b0128c2a9ec1"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L",
"type": "CVSS_V3"
}
]
}
GHSA-VR9F-9RG5-QV8M
Vulnerability from github – Published: 2025-10-01 12:30 – Updated: 2026-01-16 21:30In the Linux kernel, the following vulnerability has been resolved:
scsi: lpfc: Fix null ndlp ptr dereference in abnormal exit path for GFT_ID
An error case exit from lpfc_cmpl_ct_cmd_gft_id() results in a call to lpfc_nlp_put() with a null pointer to a nodelist structure.
Changed lpfc_cmpl_ct_cmd_gft_id() to initialize nodelist pointer upon entry.
{
"affected": [],
"aliases": [
"CVE-2022-50467"
],
"database_specific": {
"cwe_ids": [
"CWE-476"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2025-10-01T12:15:40Z",
"severity": "MODERATE"
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\nscsi: lpfc: Fix null ndlp ptr dereference in abnormal exit path for GFT_ID\n\nAn error case exit from lpfc_cmpl_ct_cmd_gft_id() results in a call to\nlpfc_nlp_put() with a null pointer to a nodelist structure.\n\nChanged lpfc_cmpl_ct_cmd_gft_id() to initialize nodelist pointer upon\nentry.",
"id": "GHSA-vr9f-9rg5-qv8m",
"modified": "2026-01-16T21:30:29Z",
"published": "2025-10-01T12:30:29Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2022-50467"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/04e7cd8c85636a329d1a6e5a269a7c8b6f71c41c"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/59b7e210a522b836a01516c71ee85d1d92c1f075"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/82dc1fe4324e2c897f2ed1c66f4fcff03094ac3a"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
]
}
GHSA-VRCC-3CH7-HPPW
Vulnerability from github – Published: 2026-10-15 20:20 – Updated: 2026-10-15 20:20Summary
The fix for GHSA-p6wj-qrr4-pgh5 addressed a nil-pointer dereference that occurs when an OpenAPI 3.1 array schema omits the items keyword (which is legal under JSON Schema 2020-12). That fix patched three decoder paths: the standalone parseArray function, the top-level array check in UrlencodedBodyDecoder, and MultipartBodyDecoder.
Three additional crash paths were missed. All of them reach the unguarded urlValuesDecoder.parseArray method (openapi3filter/req_resp_decoder.go:576) or the unguarded array branch in buildResObj (line 1015):
- Query parameters with
type: arrayand noitems. - deepObject query parameters containing an array property with no
items. - URL-encoded request bodies where the array lives inside an
allOf/anyOf/oneOfsub-schema. This path bypasses the guard added by the GHSA-p6wj-qrr4-pgh5 fix, because that guard only inspects top-level properties.
Each path is triggerable by an unauthenticated client and produces a deterministic nil-pointer panic during request validation.
Details
Root Cause
The three paths patched for GHSA-p6wj-qrr4-pgh5 each received a nil-items guard:
// standalone parseArray (line 1143) -- GUARDED
func parseArray(raw []string, schemaRef *openapi3.SchemaRef) ([]any, error) {
if schemaRef.Value.Items == nil || schemaRef.Value.Items.Value == nil { ... }
// UrlencodedBodyDecoder top-level property loop (line 1441) -- GUARDED
if propSchema.Value.Items == nil || propSchema.Value.Items.Value == nil { ... }
// MultipartBodyDecoder (line 1587) -- GUARDED
if valueSchema.Value.Items == nil { ... }
But urlValuesDecoder.parseArray (line 576), which handles query-parameter and urlencoded-body array decoding, has no such guard:
func (d *urlValuesDecoder) parseArray(raw []string, schemaRef *openapi3.SchemaRef) ([]any, error) {
var value []any
for i, v := range raw {
item, err := d.parseValue(v, schemaRef.Value.Items) // Items is nil -> panic in parseValue
parseValue immediately dereferences the schema it is handed:
func (d *urlValuesDecoder) parseValue(v string, schema *openapi3.SchemaRef) (any, error) {
if len(schema.Value.AllOf) > 0 { // schema == nil here -> nil pointer dereference
For the deepObject path, buildResObj (line 1015) has the same gap. Its array branch recurses with schema.Value.Items (line 1028), which can be nil, and the recursive call dereferences it at line 1015:
case schema.Value.Type.Is("array"):
// ...
for i := range arr {
r, err := buildResObj(params, mapKeys, strconv.Itoa(i), schema.Value.Items) // nil Items
For the allOf body path, UrlencodedBodyDecoder validates only top-level schema.Value.Properties (line 1429-1449). But decodeSchemaConstructs (line 1474) recursively walks AllOf / AnyOf / OneOf sub-schemas and calls decodeProperty on their properties, which never pass through the top-level guard and reach urlValuesDecoder.parseArray.
Proof of Concept
Environment
- kin-openapi v0.144.0 (cloned at tag
v0.144.0, commitf0407d5) - Go 1.26.7
- Each PoC calls
doc.Validate()first to prove the spec is legal, then drivesopenapi3filter.ValidateRequestdirectly
Path 1: Query parameter array without items
Spec (query param ids is type: array, no items), request GET /test?ids=1&ids=2:
Validate() ACCEPTED the spec (type: array with no items is legal under OAS 3.1)
Calling ValidateRequest directly (no net/http recover)...
panic: runtime error: invalid memory address or nil pointer dereference
[signal 0xc0000005 code=0x0 addr=0x20 ...]
github.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).parseValue(...)
req_resp_decoder.go:605
github.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).parseArray(...)
req_resp_decoder.go:580
github.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).DecodeArray(...)
req_resp_decoder.go:569
...
github.com/getkin/kin-openapi/openapi3filter.ValidateRequest(...)
validate_request.go:83
exit status 2
Path 2: deepObject nested array without items
Spec (deepObject param filter with array property tags, no items), request GET /test?filter[tags][0]=a:
Validate() ACCEPTED the spec
Calling ValidateRequest directly...
panic: runtime error: invalid memory address or nil pointer dereference
github.com/getkin/kin-openapi/openapi3filter.buildResObj(...)
req_resp_decoder.go:1012
github.com/getkin/kin-openapi/openapi3filter.buildResObj(...) <- recursive call with nil Items
req_resp_decoder.go:1028
github.com/getkin/kin-openapi/openapi3filter.makeObject(...)
req_resp_decoder.go:947
...
exit status 2
Path 3: URL-encoded body with array inside allOf (bypasses the existing fix)
Spec (urlencoded body, allOf sub-schema with array property tags, no items), request POST /submit body tags=a&tags=b:
Validate() ACCEPTED the spec
Calling ValidateRequest directly...
panic: runtime error: invalid memory address or nil pointer dereference
github.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).parseValue(...)
req_resp_decoder.go:605
github.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).parseArray(...)
req_resp_decoder.go:580
github.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).DecodeArray(...)
req_resp_decoder.go:569
...
github.com/getkin/kin-openapi/openapi3filter.decodeProperty(...) <- reached via decodeSchemaConstructs (allOf walker)
req_resp_decoder.go:1519
exit status 2
The decodeProperty frame confirms the array was reached by walking the allOf sub-schema, bypassing the top-level guard that the GHSA-p6wj-qrr4-pgh5 fix added.
Fix verification
Adding a single nil-items guard to urlValuesDecoder.parseArray (plus one to the buildResObj array branch) turns all three panics into clean validation errors:
poc1: Returned normally, err = parameter "ids" in query has an error: array items schema is required for decoding
poc2: Returned normally, err = parameter "filter" in query has an error: path tags: array items schema is required
poc3: Returned normally, err = <nil> (no panic; the allOf property is skipped by the decoder)
This confirms urlValuesDecoder.parseArray is the shared root cause for Paths 1 and 3, and buildResObj for Path 2.
Impact
An unauthenticated client can trigger a deterministic nil-pointer panic in request validation with a single crafted request, against any application that validates OAS 3.1 requests with openapi3filter where a spec declares an array without items.
The severity depends on the host:
| Deployment | Behavior | Availability impact |
|---|---|---|
Standard net/http server (validation runs in the handler goroutine) |
net/http recovers the panic per-connection; the connection is dropped and the process survives |
Per-request denial of service (A:L) |
Non-net/http hosts (e.g. fasthttp/connect adapters) or validation performed in an application-spawned goroutine |
The panic is not recovered; the process crashes | Full-process denial of service (A:H) |
This was verified in both configurations: under a live net/http server the process survived (the panic was logged as http: panic serving ... and the process exited 0 at end of test), while the direct-call PoCs above crashed the process with exit status 2.
{
"affected": [
{
"package": {
"ecosystem": "Go",
"name": "github.com/getkin/kin-openapi"
},
"ranges": [
{
"events": [
{
"introduced": "0.136.0"
},
{
"fixed": "0.149.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-100353"
],
"database_specific": {
"cwe_ids": [
"CWE-476"
],
"github_reviewed": true,
"github_reviewed_at": "2026-10-15T20:20:47Z",
"nvd_published_at": "2026-10-15T17:17:07Z",
"severity": "HIGH"
},
"details": "## Summary\n\nThe fix for GHSA-p6wj-qrr4-pgh5 addressed a nil-pointer dereference that occurs when an OpenAPI 3.1 array schema omits the `items` keyword (which is legal under JSON Schema 2020-12). That fix patched three decoder paths: the standalone `parseArray` function, the top-level array check in `UrlencodedBodyDecoder`, and `MultipartBodyDecoder`.\n\nThree additional crash paths were missed. All of them reach the unguarded `urlValuesDecoder.parseArray` method (`openapi3filter/req_resp_decoder.go:576`) or the unguarded array branch in `buildResObj` (line 1015):\n\n1. **Query parameters** with `type: array` and no `items`.\n2. **deepObject** query parameters containing an array property with no `items`.\n3. **URL-encoded request bodies** where the array lives inside an `allOf` / `anyOf` / `oneOf` sub-schema. This path bypasses the guard added by the GHSA-p6wj-qrr4-pgh5 fix, because that guard only inspects top-level properties.\n\nEach path is triggerable by an unauthenticated client and produces a deterministic nil-pointer panic during request validation.\n\n## Details\n### Root Cause\n\nThe three paths patched for GHSA-p6wj-qrr4-pgh5 each received a nil-`items` guard:\n\n```go\n// standalone parseArray (line 1143) -- GUARDED\nfunc parseArray(raw []string, schemaRef *openapi3.SchemaRef) ([]any, error) {\n\tif schemaRef.Value.Items == nil || schemaRef.Value.Items.Value == nil { ... }\n\n// UrlencodedBodyDecoder top-level property loop (line 1441) -- GUARDED\nif propSchema.Value.Items == nil || propSchema.Value.Items.Value == nil { ... }\n\n// MultipartBodyDecoder (line 1587) -- GUARDED\nif valueSchema.Value.Items == nil { ... }\n```\n\nBut `urlValuesDecoder.parseArray` (line 576), which handles query-parameter and urlencoded-body array decoding, has **no** such guard:\n\n```go\nfunc (d *urlValuesDecoder) parseArray(raw []string, schemaRef *openapi3.SchemaRef) ([]any, error) {\n\tvar value []any\n\tfor i, v := range raw {\n\t\titem, err := d.parseValue(v, schemaRef.Value.Items) // Items is nil -\u003e panic in parseValue\n```\n\n`parseValue` immediately dereferences the schema it is handed:\n\n```go\nfunc (d *urlValuesDecoder) parseValue(v string, schema *openapi3.SchemaRef) (any, error) {\n\tif len(schema.Value.AllOf) \u003e 0 { // schema == nil here -\u003e nil pointer dereference\n```\n\nFor the deepObject path, `buildResObj` (line 1015) has the same gap. Its array branch recurses with `schema.Value.Items` (line 1028), which can be nil, and the recursive call dereferences it at line 1015:\n\n```go\ncase schema.Value.Type.Is(\"array\"):\n\t// ...\n\tfor i := range arr {\n\t\tr, err := buildResObj(params, mapKeys, strconv.Itoa(i), schema.Value.Items) // nil Items\n```\n\nFor the allOf body path, `UrlencodedBodyDecoder` validates only top-level `schema.Value.Properties` (line 1429-1449). But `decodeSchemaConstructs` (line 1474) recursively walks `AllOf` / `AnyOf` / `OneOf` sub-schemas and calls `decodeProperty` on their properties, which never pass through the top-level guard and reach `urlValuesDecoder.parseArray`.\n\n## Proof of Concept\n\n### Environment\n\n- kin-openapi v0.144.0 (cloned at tag `v0.144.0`, commit `f0407d5`)\n- Go 1.26.7\n- Each PoC calls `doc.Validate()` first to prove the spec is legal, then drives `openapi3filter.ValidateRequest` directly\n\n### Path 1: Query parameter array without items\n\nSpec (query param `ids` is `type: array`, no `items`), request `GET /test?ids=1\u0026ids=2`:\n\n```\nValidate() ACCEPTED the spec (type: array with no items is legal under OAS 3.1)\nCalling ValidateRequest directly (no net/http recover)...\npanic: runtime error: invalid memory address or nil pointer dereference\n[signal 0xc0000005 code=0x0 addr=0x20 ...]\ngithub.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).parseValue(...)\n\treq_resp_decoder.go:605\ngithub.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).parseArray(...)\n\treq_resp_decoder.go:580\ngithub.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).DecodeArray(...)\n\treq_resp_decoder.go:569\n...\ngithub.com/getkin/kin-openapi/openapi3filter.ValidateRequest(...)\n\tvalidate_request.go:83\nexit status 2\n```\n\n### Path 2: deepObject nested array without items\n\nSpec (deepObject param `filter` with array property `tags`, no `items`), request `GET /test?filter[tags][0]=a`:\n\n```\nValidate() ACCEPTED the spec\nCalling ValidateRequest directly...\npanic: runtime error: invalid memory address or nil pointer dereference\ngithub.com/getkin/kin-openapi/openapi3filter.buildResObj(...)\n\treq_resp_decoder.go:1012\ngithub.com/getkin/kin-openapi/openapi3filter.buildResObj(...) \u003c- recursive call with nil Items\n\treq_resp_decoder.go:1028\ngithub.com/getkin/kin-openapi/openapi3filter.makeObject(...)\n\treq_resp_decoder.go:947\n...\nexit status 2\n```\n\n### Path 3: URL-encoded body with array inside allOf (bypasses the existing fix)\n\nSpec (urlencoded body, `allOf` sub-schema with array property `tags`, no `items`), request `POST /submit` body `tags=a\u0026tags=b`:\n\n```\nValidate() ACCEPTED the spec\nCalling ValidateRequest directly...\npanic: runtime error: invalid memory address or nil pointer dereference\ngithub.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).parseValue(...)\n\treq_resp_decoder.go:605\ngithub.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).parseArray(...)\n\treq_resp_decoder.go:580\ngithub.com/getkin/kin-openapi/openapi3filter.(*urlValuesDecoder).DecodeArray(...)\n\treq_resp_decoder.go:569\n...\ngithub.com/getkin/kin-openapi/openapi3filter.decodeProperty(...) \u003c- reached via decodeSchemaConstructs (allOf walker)\n\treq_resp_decoder.go:1519\nexit status 2\n```\n\nThe `decodeProperty` frame confirms the array was reached by walking the `allOf` sub-schema, bypassing the top-level guard that the GHSA-p6wj-qrr4-pgh5 fix added.\n\n### Fix verification\n\nAdding a single nil-`items` guard to `urlValuesDecoder.parseArray` (plus one to the `buildResObj` array branch) turns all three panics into clean validation errors:\n\n```\npoc1: Returned normally, err = parameter \"ids\" in query has an error: array items schema is required for decoding\npoc2: Returned normally, err = parameter \"filter\" in query has an error: path tags: array items schema is required\npoc3: Returned normally, err = \u003cnil\u003e (no panic; the allOf property is skipped by the decoder)\n```\n\nThis confirms `urlValuesDecoder.parseArray` is the shared root cause for Paths 1 and 3, and `buildResObj` for Path 2.\n\n## Impact\n\nAn unauthenticated client can trigger a deterministic nil-pointer panic in request validation with a single crafted request, against any application that validates OAS 3.1 requests with `openapi3filter` where a spec declares an array without `items`.\n\nThe severity depends on the host:\n\n| Deployment | Behavior | Availability impact |\n|------------|----------|---------------------|\n| Standard `net/http` server (validation runs in the handler goroutine) | `net/http` recovers the panic per-connection; the connection is dropped and the process survives | Per-request denial of service (A:L) |\n| Non-`net/http` hosts (e.g. fasthttp/connect adapters) or validation performed in an application-spawned goroutine | The panic is not recovered; the process crashes | Full-process denial of service (A:H) |\n\nThis was verified in both configurations: under a live `net/http` server the process survived (the panic was logged as `http: panic serving ...` and the process exited 0 at end of test), while the direct-call PoCs above crashed the process with `exit status 2`.",
"id": "GHSA-vrcc-3ch7-hppw",
"modified": "2026-10-15T20:20:47Z",
"published": "2026-10-15T20:20:47Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/getkin/kin-openapi/security/advisories/GHSA-vrcc-3ch7-hppw"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-100353"
},
{
"type": "WEB",
"url": "https://github.com/getkin/kin-openapi/commit/9814d1f2b212f3508bf52296aa9e248b2122a5a9"
},
{
"type": "PACKAGE",
"url": "https://github.com/getkin/kin-openapi"
},
{
"type": "WEB",
"url": "https://github.com/getkin/kin-openapi/releases/tag/v0.149.0"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
],
"summary": "kin-openapi: Incomplete Fix for GHSA-p6wj-qrr4-pgh5: Three Unpatched nil-Items Panic Paths in openapi3filter"
}
Mitigation MIT-56
For any pointers that could have been modified or provided from a function that can return NULL, check the pointer for NULL before use. When working with a multithreaded or otherwise asynchronous environment, ensure that proper locking APIs are used to lock before the check, and unlock when it has finished [REF-1484].
Mitigation
Select a programming language that is not susceptible to these issues.
Mitigation
Check the results of all functions that return a value and verify that the value is non-null before acting upon it.
Mitigation
Identify all variables and data stores that receive information from external sources, and apply input validation to make sure that they are only initialized to expected values.
Mitigation
Explicitly initialize all variables and other data stores, either during declaration or just before the first usage.
No CAPEC attack patterns related to this CWE.