CVE-2022-49744

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

mm/uffd: fix pte marker when fork() without fork event

Patch series "mm: Fixes on pte markers".

Patch 1 resolves the syzkiller report from Pengfei.

Patch 2 further harden pte markers when used with the recent swapin error markers. The major case is we should persist a swapin error marker after fork(), so child shouldn't read a corrupted page.

This patch (of 2):

When fork(), dst_vma is not guaranteed to have VM_UFFD_WP even if src may have it and has pte marker installed. The warning is improper along with the comment. The right thing is to inherit the pte marker when needed, or keep the dst pte empty.

A vague guess is this happened by an accident when there's the prior patch to introduce src/dst vma into this helper during the uffd-wp feature got developed and I probably messed up in the rebase, since if we replace dst_vma with src_vma the warning & comment it all makes sense too.

Hugetlb did exactly the right here (copy_hugetlb_page_range()). Fix the general path.

Reproducer:

https://github.com/xupengfe/syzkaller_logs/blob/main/221208_115556_copy_page_range/repro.c

Bugzilla report: https://bugzilla.kernel.org/show_bug.cgi?id=216808

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

mm/uffd: fix pte marker when fork() without fork event

Patch series "mm: Fixes on pte markers".

Patch 1 resolves the syzkiller report from Pengfei.

Patch 2 further harden pte markers when used with the recent swapin error markers. The major case is we should persist a swapin error marker after fork(), so child shouldn't read a corrupted page.

This patch (of 2):

When fork(), dst_vma is not guaranteed to have VM_UFFD_WP even if src may have it and has pte marker installed. The warning is improper along with the comment. The right thing is to inherit the pte marker when needed, or keep the dst pte empty.

A vague guess is this happened by an accident when there's the prior patch to introduce src/dst vma into this helper during the uffd-wp feature got developed and I probably messed up in the rebase, since if we replace dst_vma with src_vma the warning & comment it all makes sense too.

Hugetlb did exactly the right here (copy_hugetlb_page_range()). Fix the general path.

Reproducer:

https://github.com/xupengfe/syzkaller_logs/blob/main/221208_115556_copy_page_range/repro.c

Bugzilla report: https://bugzilla.kernel.org/show_bug.cgi?id=216808

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

mm/uffd: fix pte marker when fork() without fork event

Patch series "mm: Fixes on pte markers".

Patch 1 resolves the syzkiller report from Pengfei.

Patch 2 further harden pte markers when used with the recent swapin error markers. The major case is we should persist a swapin error marker after fork(), so child shouldn't read a corrupted page.

This patch (of 2):

When fork(), dst_vma is not guaranteed to have VM_UFFD_WP even if src may have it and has pte marker installed. The warning is improper along with the comment. The right thing is to inherit the pte marker when needed, or keep the dst pte empty.

A vague guess is this happened by an accident when there's the prior patch to introduce src/dst vma into this helper during the uffd-wp feature got developed and I probably messed up in the rebase, since if we replace dst_vma with src_vma the warning & comment it all makes sense too.

Hugetlb did exactly the right here (copy_hugetlb_page_range()). Fix the general path.

Reproducer:

https://github.com/xupengfe/syzkaller_logs/blob/main/221208_115556_copy_page_range/repro.c

Bugzilla report: https://bugzilla.kernel.org/show_bug.cgi?id=216808

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

mm/uffd: fix pte marker when fork() without fork event

Patch series "mm: Fixes on pte markers".

Patch 1 resolves the syzkiller report from Pengfei.

Patch 2 further harden pte markers when used with the recent swapin error markers. The major case is we should persist a swapin error marker after fork(), so child shouldn't read a corrupted page.

This patch (of 2):

When fork(), dst_vma is not guaranteed to have VM_UFFD_WP even if src may have it and has pte marker installed. The warning is improper along with the comment. The right thing is to inherit the pte marker when needed, or keep the dst pte empty.

A vague guess is this happened by an accident when there's the prior patch to introduce src/dst vma into this helper during the uffd-wp feature got developed and I probably messed up in the rebase, since if we replace dst_vma with src_vma the warning & comment it all makes sense too.

Hugetlb did exactly the right here (copy_hugetlb_page_range()). Fix the general path.

Reproducer:

https://github.com/xupengfe/syzkaller_logs/blob/main/221208_115556_copy_page_range/repro.c

Bugzilla report: https://bugzilla.kernel.org/show_bug.cgi?id=216808

In the Linux kernel, the following vulnerability has been resolved: mm/uffd: fix pte marker when fork() without fork event Patch series "mm: Fixes on pte markers". Patch 1 resolves the syzkiller report from Pengfei. Patch 2 further harden pte markers when used with the recent swapin error markers. The major case is we should persist a swapin error marker after fork(), so child shouldn't read a corrupted page. This patch (of 2): When fork(), dst_vma is not guaranteed to have VM_UFFD_WP even if src may have it and has pte marker installed. The warning is improper along with the comment. The right thing is to inherit the pte marker when needed, or keep the dst pte empty. A vague guess is this happened by an accident when there's the prior patch to introduce src/dst vma into this helper during the uffd-wp feature got developed and I probably messed up in the rebase, since if we replace dst_vma with src_vma the warning & comment it all makes sense too. Hugetlb did exactly the right here (copy_hugetlb_page_range()). Fix the general path. Reproducer: https://github.com/xupengfe/syzkaller_logs/blob/main/221208_115556_copy_page_range/repro.c Bugzilla report: https://bugzilla.kernel.org/show_bug.cgi?id=216808

• net/sched: fix pedit partial COW leading to page cache corruption {CVE-2026-46331}
• unshare: fix unshare_fs() handling {CVE-2026-43472}
• net/sched: sch_hfsc: fix divide-by-zero in rtsc_min() {CVE-2026-31423}
• net/sched: cls_flow: fix NULL pointer dereference on shared blocks {CVE-2026-31422}
• net/sched: cls_fw: fix NULL pointer dereference on shared blocks {CVE-2026-31421}
• dm: fix queue start/stop imbalance under suspend/load/resume races {CVE-2025-40134}
• ext4: avoid allocating blocks from corrupted group in ext4_mb_try_best_found() {CVE-2024-26773}
• ext4: fix i_data_sem unlock order in ext4_ind_migrate() {CVE-2024-50006}
• xen: Fix the issue of resource not being properly released in xenbus_dev_probe() {CVE-2024-53198}
• HID: pidff: Fix null pointer dereference in pidff_find_fields {CVE-2025-37862}
• crypto: af_alg - Set merge to zero early in af_alg_sendmsg {CVE-2025-39931}
• x86/cpu/hygon: Add missing resctrl_cpu_detect() in bsp_init helper {CVE-2025-39681}
• can: kvaser_usb: kvaser_usb_read_bulk_callback(): fix URB memory leak {CVE-2026-23061}
• bonding: restore IFF_MASTER/SLAVE flags on bond enslave ether type change {CVE-2023-53103}
• NFS: Fix a potential data corruption {CVE-2023-53711}
• NFS: Fix a use after free in nfs_direct_join_group() {CVE-2023-53711}
• hwmon: (w83791d) Fix NULL pointer dereference by removing unnecessary structure field {CVE-2021-47386}
• scsi: bnx2fc: Remove spin_lock_bh while releasing resources after upload {CVE-2024-36919}
• x86/mm: Ensure input to pfn_to_kaddr() is treated as a 64-bit type {CVE-2023-52659}
• Input: uinput - reject requests with unreasonable number of slots {CVE-2024-46745}
• afs: Fix the maximum cell name length {CVE-2025-21646}
• drm/i915: Avoid potential vm use-after-free {CVE-2023-52931}
• interconnect: fix mem leak when freeing nodes {CVE-2023-53096}
• drm/vkms: Fix memory leak in vkms_init() {CVE-2022-50269}
• USB: dwc3: fix memory leak with using debugfs_lookup() {CVE-2023-53415}
• mm/uffd: fix pte marker when fork() without fork event {CVE-2022-49744}
• scsi: qla2xxx: Fix memory leak in qla2x00_probe_one() {CVE-2023-53696}
• ipmi_si: fix a memleak in try_smi_init() {CVE-2023-53611}
• KVM: Explicitly verify target vCPU is online in kvm_get_vcpu() {CVE-2024-58083}
• scsi: lpfc: Release hbalock before calling lpfc_worker_wake_up() {CVE-2024-36924}
• firmware: cs_dsp: Validate payload length before processing block {CVE-2024-42237}
• hwmon: (w83792d) Fix NULL pointer dereference by removing unnecessary structure field {CVE-2021-47385}
• ftrace: Avoid potential division by zero in function_stat_show() {CVE-2025-21898}
• IB/core: Implement a limit on UMAD receive List {CVE-2024-42145}
• vfio/pci: Disable auto-enable of exclusive INTx IRQ {CVE-2024-27437}
• netfilter: nf_tables: reject duplicate device on updates {CVE-2025-38678}
• PCI: vmd: Make vmd_dev::cfg_lock a raw_spinlock_t type {CVE-2025-23161}
• net: nexthop: Initialize all fields in dumped nexthops {CVE-2024-42283}
• tracing: Make sure trace_printk() can output as soon as it can be used {CVE-2023-53007}
• net: tls: explicitly disallow disconnect {CVE-2025-37756}
• KVM: x86: Reject Hyper-V's SEND_IPI hypercalls if local APIC isn't in-kernel {CVE-2025-21779}
• ext4: turn quotas off if mount failed after enabling quotas {CVE-2023-54153}
• NFC: digital: Bounds check NFC-A cascade depth in SDD response handler {CVE-2026-31622}
• block: fix overflow in blk_ioctl_discard() {CVE-2024-36917}
• usb: xhci: Check endpoint is valid before dereferencing it {CVE-2023-52901}
• iio: core: fix memleak in iio_device_register_sysfs {CVE-2023-52643}
• platform/x86/amd/pmc: Detect when STB is not available {CVE-2024-53072}
• mm: memcg: fix NULL pointer in mem_cgroup_track_foreign_dirty_slowpath() {CVE-2023-52939}
• dm thin: make get_first_thin use rcu-safe list first function {CVE-2025-21664}
• bus: mhi: host: pci_generic: Use pci_try_reset_function() to avoid deadlock {CVE-2025-21951}
• bonding: provide a net pointer to…

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

mm/uffd: fix pte marker when fork() without fork event

Patch series "mm: Fixes on pte markers".

Patch 1 resolves the syzkiller report from Pengfei.

Patch 2 further harden pte markers when used with the recent swapin error markers. The major case is we should persist a swapin error marker after fork(), so child shouldn't read a corrupted page.

This patch (of 2):

When fork(), dst_vma is not guaranteed to have VM_UFFD_WP even if src may have it and has pte marker installed. The warning is improper along with the comment. The right thing is to inherit the pte marker when needed, or keep the dst pte empty.

A vague guess is this happened by an accident when there's the prior patch to introduce src/dst vma into this helper during the uffd-wp feature got developed and I probably messed up in the rebase, since if we replace dst_vma with src_vma the warning & comment it all makes sense too.

Hugetlb did exactly the right here (copy_hugetlb_page_range()). Fix the general path.

Reproducer:

https://github.com/xupengfe/syzkaller_logs/blob/main/221208_115556_copy_page_range/repro.c

Bugzilla report: https://bugzilla.kernel.org/show_bug.cgi?id=216808

In the Linux kernel, the following vulnerability has been resolved: mm/uffd: fix pte marker when fork() without fork event Patch series "mm: Fixes on pte markers". Patch 1 resolves the syzkiller report from Pengfei. Patch 2 further harden pte markers when used with the recent swapin error markers. The major case is we should persist a swapin error marker after fork(), so child shouldn't read a corrupted page. This patch (of 2): When fork(), dst_vma is not guaranteed to have VM_UFFD_WP even if src may have it and has pte marker installed. The warning is improper along with the comment. The right thing is to inherit the pte marker when needed, or keep the dst pte empty. A vague guess is this happened by an accident when there's the prior patch to introduce src/dst vma into this helper during the uffd-wp feature got developed and I probably messed up in the rebase, since if we replace dst_vma with src_vma the warning & comment it all makes sense too. Hugetlb did exactly the right here (copy_hugetlb_page_range()). Fix the general path. Reproducer: https://github.com/xupengfe/syzkaller_logs/blob/main/221208_115556_copy_page_range/repro.c Bugzilla report: https://bugzilla.kernel.org/show_bug.cgi?id=216808