CVE-2022-49882

In the Linux kernel, the following vulnerability has been resolved:

KVM: Reject attempts to consume or refresh inactive gfn_to_pfn_cache

Reject kvm_gpc_check() and kvm_gpc_refresh() if the cache is inactive. Not checking the active flag during refresh is particularly egregious, as KVM can end up with a valid, inactive cache, which can lead to a variety of use-after-free bugs, e.g. consuming a NULL kernel pointer or missing an mmu_notifier invalidation due to the cache not being on the list of gfns to invalidate.

Note, "active" needs to be set if and only if the cache is on the list of caches, i.e. is reachable via mmu_notifier events. If a relevant mmu_notifier event occurs while the cache is "active" but not on the list, KVM will not acquire the cache's lock and so will not serailize the mmu_notifier event with active users and/or kvm_gpc_refresh().

A race between KVM_XEN_ATTR_TYPE_SHARED_INFO and KVM_XEN_HVM_EVTCHN_SEND can be exploited to trigger the bug.

1. Deactivate shinfo cache:

kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO kvm_gpc_deactivate kvm_gpc_unmap gpc->valid = false gpc->khva = NULL gpc->active = false

Result: active = false, valid = false

2. Cause cache refresh:

kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast kvm_gpc_check return -EWOULDBLOCK because !gpc->valid kvm_xen_set_evtchn_fast return -EWOULDBLOCK kvm_gpc_refresh hva_to_pfn_retry gpc->valid = true gpc->khva = not NULL

Result: active = false, valid = true

3. Race ioctl KVM_XEN_HVM_EVTCHN_SEND against ioctl KVM_XEN_ATTR_TYPE_SHARED_INFO:

kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast read_lock gpc->lock kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO mutex_lock kvm->lock kvm_xen_shared_info_init kvm_gpc_activate gpc->khva = NULL kvm_gpc_check [ Check passes because gpc->valid is still true, even though gpc->khva is already NULL. ] shinfo = gpc->khva pending_bits = shinfo->evtchn_pending CRASH: test_and_set_bit(..., pending_bits)

In the Linux kernel, the following vulnerability has been resolved:

KVM: Reject attempts to consume or refresh inactive gfn_to_pfn_cache

Reject kvm_gpc_check() and kvm_gpc_refresh() if the cache is inactive. Not checking the active flag during refresh is particularly egregious, as KVM can end up with a valid, inactive cache, which can lead to a variety of use-after-free bugs, e.g. consuming a NULL kernel pointer or missing an mmu_notifier invalidation due to the cache not being on the list of gfns to invalidate.

Note, "active" needs to be set if and only if the cache is on the list of caches, i.e. is reachable via mmu_notifier events. If a relevant mmu_notifier event occurs while the cache is "active" but not on the list, KVM will not acquire the cache's lock and so will not serailize the mmu_notifier event with active users and/or kvm_gpc_refresh().

A race between KVM_XEN_ATTR_TYPE_SHARED_INFO and KVM_XEN_HVM_EVTCHN_SEND can be exploited to trigger the bug.

1. Deactivate shinfo cache:

kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO kvm_gpc_deactivate kvm_gpc_unmap gpc->valid = false gpc->khva = NULL gpc->active = false

Result: active = false, valid = false

2. Cause cache refresh:

kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast kvm_gpc_check return -EWOULDBLOCK because !gpc->valid kvm_xen_set_evtchn_fast return -EWOULDBLOCK kvm_gpc_refresh hva_to_pfn_retry gpc->valid = true gpc->khva = not NULL

Result: active = false, valid = true

3. Race ioctl KVM_XEN_HVM_EVTCHN_SEND against ioctl KVM_XEN_ATTR_TYPE_SHARED_INFO:

kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast read_lock gpc->lock kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO mutex_lock kvm->lock kvm_xen_shared_info_init kvm_gpc_activate gpc->khva = NULL kvm_gpc_check [ Check passes because gpc->valid is still true, even though gpc->khva is already NULL. ] shinfo = gpc->khva pending_bits = shinfo->evtchn_pending CRASH: test_and_set_bit(..., pending_bits)

In the Linux kernel, the following vulnerability has been resolved:

KVM: Reject attempts to consume or refresh inactive gfn_to_pfn_cache

Reject kvm_gpc_check() and kvm_gpc_refresh() if the cache is inactive. Not checking the active flag during refresh is particularly egregious, as KVM can end up with a valid, inactive cache, which can lead to a variety of use-after-free bugs, e.g. consuming a NULL kernel pointer or missing an mmu_notifier invalidation due to the cache not being on the list of gfns to invalidate.

Note, "active" needs to be set if and only if the cache is on the list of caches, i.e. is reachable via mmu_notifier events. If a relevant mmu_notifier event occurs while the cache is "active" but not on the list, KVM will not acquire the cache's lock and so will not serailize the mmu_notifier event with active users and/or kvm_gpc_refresh().

A race between KVM_XEN_ATTR_TYPE_SHARED_INFO and KVM_XEN_HVM_EVTCHN_SEND can be exploited to trigger the bug.

1. Deactivate shinfo cache:

kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO kvm_gpc_deactivate kvm_gpc_unmap gpc->valid = false gpc->khva = NULL gpc->active = false

Result: active = false, valid = false

2. Cause cache refresh:

kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast kvm_gpc_check return -EWOULDBLOCK because !gpc->valid kvm_xen_set_evtchn_fast return -EWOULDBLOCK kvm_gpc_refresh hva_to_pfn_retry gpc->valid = true gpc->khva = not NULL

Result: active = false, valid = true

3. Race ioctl KVM_XEN_HVM_EVTCHN_SEND against ioctl KVM_XEN_ATTR_TYPE_SHARED_INFO:

kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast read_lock gpc->lock kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO mutex_lock kvm->lock kvm_xen_shared_info_init kvm_gpc_activate gpc->khva = NULL kvm_gpc_check [ Check passes because gpc->valid is still true, even though gpc->khva is already NULL. ] shinfo = gpc->khva pending_bits = shinfo->evtchn_pending CRASH: test_and_set_bit(..., pending_bits)

In the Linux kernel, the following vulnerability has been resolved:

KVM: Reject attempts to consume or refresh inactive gfn_to_pfn_cache

Reject kvm_gpc_check() and kvm_gpc_refresh() if the cache is inactive. Not checking the active flag during refresh is particularly egregious, as KVM can end up with a valid, inactive cache, which can lead to a variety of use-after-free bugs, e.g. consuming a NULL kernel pointer or missing an mmu_notifier invalidation due to the cache not being on the list of gfns to invalidate.

Note, "active" needs to be set if and only if the cache is on the list of caches, i.e. is reachable via mmu_notifier events. If a relevant mmu_notifier event occurs while the cache is "active" but not on the list, KVM will not acquire the cache's lock and so will not serailize the mmu_notifier event with active users and/or kvm_gpc_refresh().

A race between KVM_XEN_ATTR_TYPE_SHARED_INFO and KVM_XEN_HVM_EVTCHN_SEND can be exploited to trigger the bug.

1. Deactivate shinfo cache:

kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO kvm_gpc_deactivate kvm_gpc_unmap gpc->valid = false gpc->khva = NULL gpc->active = false

Result: active = false, valid = false

2. Cause cache refresh:

kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast kvm_gpc_check return -EWOULDBLOCK because !gpc->valid kvm_xen_set_evtchn_fast return -EWOULDBLOCK kvm_gpc_refresh hva_to_pfn_retry gpc->valid = true gpc->khva = not NULL

Result: active = false, valid = true

3. Race ioctl KVM_XEN_HVM_EVTCHN_SEND against ioctl KVM_XEN_ATTR_TYPE_SHARED_INFO:

kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast read_lock gpc->lock kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO mutex_lock kvm->lock kvm_xen_shared_info_init kvm_gpc_activate gpc->khva = NULL kvm_gpc_check [ Check passes because gpc->valid is still true, even though gpc->khva is already NULL. ] shinfo = gpc->khva pending_bits = shinfo->evtchn_pending CRASH: test_and_set_bit(..., pending_bits)

In the Linux kernel, the following vulnerability has been resolved:

KVM: Reject attempts to consume or refresh inactive gfn_to_pfn_cache

Reject kvm_gpc_check() and kvm_gpc_refresh() if the cache is inactive. Not checking the active flag during refresh is particularly egregious, as KVM can end up with a valid, inactive cache, which can lead to a variety of use-after-free bugs, e.g. consuming a NULL kernel pointer or missing an mmu_notifier invalidation due to the cache not being on the list of gfns to invalidate.

Note, "active" needs to be set if and only if the cache is on the list of caches, i.e. is reachable via mmu_notifier events. If a relevant mmu_notifier event occurs while the cache is "active" but not on the list, KVM will not acquire the cache's lock and so will not serailize the mmu_notifier event with active users and/or kvm_gpc_refresh().

A race between KVM_XEN_ATTR_TYPE_SHARED_INFO and KVM_XEN_HVM_EVTCHN_SEND can be exploited to trigger the bug.

1. Deactivate shinfo cache:

kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO kvm_gpc_deactivate kvm_gpc_unmap gpc->valid = false gpc->khva = NULL gpc->active = false

Result: active = false, valid = false

2. Cause cache refresh:

kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast kvm_gpc_check return -EWOULDBLOCK because !gpc->valid kvm_xen_set_evtchn_fast return -EWOULDBLOCK kvm_gpc_refresh hva_to_pfn_retry gpc->valid = true gpc->khva = not NULL

Result: active = false, valid = true

3. Race ioctl KVM_XEN_HVM_EVTCHN_SEND against ioctl KVM_XEN_ATTR_TYPE_SHARED_INFO:

kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast read_lock gpc->lock kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO mutex_lock kvm->lock kvm_xen_shared_info_init kvm_gpc_activate gpc->khva = NULL kvm_gpc_check [ Check passes because gpc->valid is still true, even though gpc->khva is already NULL. ] shinfo = gpc->khva pending_bits = shinfo->evtchn_pending CRASH: test_and_set_bit(..., pending_bits)

In the Linux kernel, the following vulnerability has been resolved: KVM: Reject attempts to consume or refresh inactive gfn_to_pfn_cache Reject kvm_gpc_check() and kvm_gpc_refresh() if the cache is inactive. Not checking the active flag during refresh is particularly egregious, as KVM can end up with a valid, inactive cache, which can lead to a variety of use-after-free bugs, e.g. consuming a NULL kernel pointer or missing an mmu_notifier invalidation due to the cache not being on the list of gfns to invalidate. Note, "active" needs to be set if and only if the cache is on the list of caches, i.e. is reachable via mmu_notifier events. If a relevant mmu_notifier event occurs while the cache is "active" but not on the list, KVM will not acquire the cache's lock and so will not serailize the mmu_notifier event with active users and/or kvm_gpc_refresh(). A race between KVM_XEN_ATTR_TYPE_SHARED_INFO and KVM_XEN_HVM_EVTCHN_SEND can be exploited to trigger the bug. 1. Deactivate shinfo cache: kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO kvm_gpc_deactivate kvm_gpc_unmap gpc->valid = false gpc->khva = NULL gpc->active = false Result: active = false, valid = false 2. Cause cache refresh: kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast kvm_gpc_check return -EWOULDBLOCK because !gpc->valid kvm_xen_set_evtchn_fast return -EWOULDBLOCK kvm_gpc_refresh hva_to_pfn_retry gpc->valid = true gpc->khva = not NULL Result: active = false, valid = true 3. Race ioctl KVM_XEN_HVM_EVTCHN_SEND against ioctl KVM_XEN_ATTR_TYPE_SHARED_INFO: kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast read_lock gpc->lock kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO mutex_lock kvm->lock kvm_xen_shared_info_init kvm_gpc_activate gpc->khva = NULL kvm_gpc_check [ Check passes because gpc->valid is still true, even though gpc->khva is already NULL. ] shinfo = gpc->khva pending_bits = shinfo->evtchn_pending CRASH: test_and_set_bit(..., pending_bits)

• Bluetooth: MGMT: Protect mgmt_pending list with its own lock {CVE-2025-38117}
• padata: Fix pd UAF once and for all {CVE-2025-38584}
• wifi: cfg80211: Fix use after free for wext {CVE-2023-53153}
• padata: Fix list iterator in padata_do_serial()
• padata: do not leak refcount in reorder_work {CVE-2025-38031}
• padata: avoid UAF for reorder_work {CVE-2025-21726}
• padata: add pd get/put refcnt helper
• net/smc: fix UAF on smcsk after smc_listen_out() {CVE-2025-38734}
• scsi: mpi3mr: Bad drive in topology results kernel crash {CVE-2023-53037}
• scsi: mpi3mr: Driver unload crashes host when enhanced logging is enabled
• wifi: brcmfmac: Fix potential stack-out-of-bounds in brcmf_c_preinit_dcmds() {CVE-2022-50258}
• igb: Fix igb_down hung on surprise removal {CVE-2023-53148}
• drm/i915/sseu: fix max_subslices array-index-out-of-bounds access {CVE-2023-53112}
• i40e: fix idx validation in config queues msg {CVE-2025-39971}
• dm-bufio: don't schedule in atomic context {CVE-2025-37928}
• fs: prevent out-of-bounds array speculation when closing a file descriptor {CVE-2023-53117}
• vhost-scsi: Fix handling of multiple calls to vhost_scsi_set_endpoint {CVE-2025-22083}
• iommu/amd: Fix potential buffer overflow in parse_ivrs_acpihid {CVE-2025-37927}
• tracing: Fix oob write in trace_seq_to_buffer() {CVE-2025-37923}
• KVM: x86: Reset IRTE to host control if new route isn't postable {CVE-2025-37885}
• KVM: arm64: Tear down vGIC on failed vCPU creation {CVE-2025-37849}
• KVM: Reject attempts to consume or refresh inactive gfn_to_pfn_cache {CVE-2022-49882}
• jbd2: remove wrong sb->s_sequence check {CVE-2025-37839}
• power: supply: gpio-charger: Fix set charge current limits {CVE-2024-57792}
• ASoC: dapm: Fix UAF for snd_soc_pcm_runtime object {CVE-2024-46798}
• ipv6: prevent possible UAF in ip6_xmit() {CVE-2024-44985}
• ipv6: fix possible UAF in ip6_finish_output2() {CVE-2024-44986}
• netfilter: flowtable: validate vlan header {CVE-2024-44983}
• usb: typec: fix potential out of bounds in ucsi_ccg_update_set_new_cam_cmd() {CVE-2024-50268}
• usb: vhci-hcd: Do not drop references before new references are gained {CVE-2024-43883}
• drm/amd/display: Fix slab-use-after-free on hdcp_work {CVE-2025-21968}
• drm/amd/display: Check num_valid_sets before accessing reader_wm_sets[] {CVE-2024-46815}
• drm/amd/display: Check index msg_id before read or write {CVE-2024-42121}
• drm/amd/display: Check pipe offset before setting vblank {CVE-2024-42120}
• drm/amdkfd: don't allow mapping the MMIO HDP page with large pages {CVE-2024-41011}
• drm/amdkfd: Fix an illegal memory access {CVE-2023-53090}
• drm/shmem-helper: Remove another errant put in error path {CVE-2023-53084}
• drm/amd/display: fix shift-out-of-bounds in CalculateVMAndRowBytes {CVE-2023-53077}
• drm/i915: mark requests for GuC virtual engines to avoid use-after-free {CVE-2023-53552}
• ptp: fix integer overflow in max_vclocks_store {CVE-2024-40994}
• nvmet: avoid potential UAF in nvmet_req_complete() {CVE-2023-53116}
• loop: Fix use-after-free issues {CVE-2023-53111}
• ftrace: Fix invalid address access in lookup_rec() when index is 0 {CVE-2023-53075}
• mptcp: use the workqueue to destroy unaccepted sockets {CVE-2023-53072}
• perf/core: Fix perf_output_begin parameter is incorrectly invoked in perf_event_bpf_output {CVE-2023-53065}
• erspan: do not use skb_mac_header() in ndo_start_xmit() {CVE-2023-53053}
• HID: intel-ish-hid: ipc: Fix potential use-after-free in work function {CVE-2023-53039}
• netlink: prevent potential spectre v1 gadgets {CVE-2023-53000}
• coresight: Clear the connection field properly {CVE-2022-50214}
• Bluetooth: btsdio: fix use after free bug in btsdio_remove due to race condition {CVE-2023-53145}
• Bluetooth: hci_event: Fix UAF in hci_acl_create_conn_sync {CVE-2025-39982}
• x86/tdx: Fix "in-kernel MMIO" check {CVE-2024-47727}
• arm64: sme: Use STR P to clear FFR context field in streaming SVE mode {CVE-2023-53713}
• wifi: mt76: mt7921: fix error code of return in…

In the Linux kernel, the following vulnerability has been resolved: KVM: Reject attempts to consume or refresh inactive gfn_to_pfn_cache Reject kvm_gpc_check() and kvm_gpc_refresh() if the cache is inactive. Not checking the active flag during refresh is particularly egregious, as KVM can end up with a valid, inactive cache, which can lead to a variety of use-after-free bugs, e.g. consuming a NULL kernel pointer or missing an mmu_notifier invalidation due to the cache not being on the list of gfns to invalidate. Note, "active" needs to be set if and only if the cache is on the list of caches, i.e. is reachable via mmu_notifier events. If a relevant mmu_notifier event occurs while the cache is "active" but not on the list, KVM will not acquire the cache's lock and so will not serailize the mmu_notifier event with active users and/or kvm_gpc_refresh(). A race between KVM_XEN_ATTR_TYPE_SHARED_INFO and KVM_XEN_HVM_EVTCHN_SEND can be exploited to trigger the bug. 1. Deactivate shinfo cache: kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO kvm_gpc_deactivate kvm_gpc_unmap gpc->valid = false gpc->khva = NULL gpc->active = false Result: active = false, valid = false 2. Cause cache refresh: kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast kvm_gpc_check return -EWOULDBLOCK because !gpc->valid kvm_xen_set_evtchn_fast return -EWOULDBLOCK kvm_gpc_refresh hva_to_pfn_retry gpc->valid = true gpc->khva = not NULL Result: active = false, valid = true 3. Race ioctl KVM_XEN_HVM_EVTCHN_SEND against ioctl KVM_XEN_ATTR_TYPE_SHARED_INFO: kvm_arch_vm_ioctl case KVM_XEN_HVM_EVTCHN_SEND kvm_xen_hvm_evtchn_send kvm_xen_set_evtchn kvm_xen_set_evtchn_fast read_lock gpc->lock kvm_xen_hvm_set_attr case KVM_XEN_ATTR_TYPE_SHARED_INFO mutex_lock kvm->lock kvm_xen_shared_info_init kvm_gpc_activate gpc->khva = NULL kvm_gpc_check [ Check passes because gpc->valid is still true, even though gpc->khva is already NULL. ] shinfo = gpc->khva pending_bits = shinfo->evtchn_pending CRASH: test_and_set_bit(..., pending_bits)