Export limit exceeded: 378141 CVEs match your query. Please refine your search to export 10,000 CVEs or fewer.
Search
Search Results (378141 CVEs found)
| CVE | Vendors | Products | Updated | CVSS v3.1 |
|---|---|---|---|---|
| CVE-2026-72193 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: ntfs3: cap RESTART_TABLE free-chain walker at rt->used A crafted NTFS3 disk image triggers an in-kernel infinite loop at mount time, hanging the mounting thread and firing the soft-lockup watchdog within ~22s on multi-CPU hosts (panic with kernel.softlockup_panic=1). The bug is reachable from desktop USB auto-mount on distributions where udisks2 routes the NTFS signature to the in-tree ntfs3 driver (Arch family and an increasing fraction of Fedora / openSUSE / RHEL deployments); CAP_SYS_ADMIN-class manual mount elsewhere. check_rstbl()'s second walker iterates the free-entry singly-linked list headed by rt->first_free with no upper bound on iteration count: for (off = ff; off;) { if (off == RESTART_ENTRY_ALLOCATED) return false; off = le32_to_cpu(*(__le32 *)Add2Ptr(rt, off)); if (off > ts - sizeof(__le32)) return false; } The existing guards cover three exits: end-of-list (off == 0), the in-use marker (off == RESTART_ENTRY_ALLOCATED), and out-of-bounds (off > ts - sizeof(__le32)). None of the three prevents an in-bounds cycle. A crafted on-disk RESTART_TABLE whose free chain contains a self-loop or A->B->A cycle whose offsets satisfy: - in range [sizeof(struct RESTART_TABLE), ts - sizeof(__le32)] - (off - sizeof(struct RESTART_TABLE)) % rsize == 0 passes all existing guards and spins the mount-time thread forever. Reproduced in UML by hand-forging a 2 MB NTFS3 image whose journal RESTART_TABLE first_free = 0x18 and whose entry at offset 0x18 stores 0x18 as its next pointer; mount of the forged image with the in-tree ntfs3 driver never returns. Bound the walker by rt->used. Each entry on a legitimate free chain is unique, and the total slot count is ne = le16_to_cpu (rt->used). A traversal that visits more than ne slots is by construction malformed; reject it as a corrupt RESTART_TABLE. After this patch, mount of the forged image returns with -EINVAL and a log_replay failure message, and mkntfs-produced legitimate images mount cleanly (verified in the same UML harness). | ||||
| CVE-2026-72192 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: ntfs3: bound to_move in indx_insert_into_root before hdr_insert_head indx_insert_into_root() promotes a full resident $INDEX_ROOT into $INDEX_ALLOCATION and copies all non-last resident root entries into a newly allocated INDEX_BUFFER via hdr_insert_head(). The source byte count 'to_move' is summed from the on-disk resident entry sizes and is independent of the destination buffer size, which comes from root->index_block_size (via indx->index_bits). A crafted NTFS image that keeps a valid, full resident root but shrinks root->index_block_size down to 512 after the root has been populated makes hdr_insert_head() memcpy attacker-controlled resident entry bytes past the end of the kmalloc(1u << indx->index_bits) allocation returned by indx_new(). For a 512-byte destination and a resident root whose non-last entries total 560 bytes, the memcpy overruns by 120 bytes and a following memmove extends the highest written offset to 136 bytes past the allocation. The overflow bytes are a direct copy of on-disk entries (via kmemdup), so they are fully attacker-controlled. The write is reachable from unprivileged open(O_CREAT) on a mounted crafted NTFS image: a single sufficiently long create in a directory whose resident root is already full forces root promotion and triggers the copy. This is a controlled out-of-bounds write of 120-136 bytes past a kmalloc(index_block_size) allocation, with attacker-controlled content. It is a bounded adjacent-heap corruption primitive; it is not an arbitrary-address write. Successful exploitation into a named victim object depends on the surrounding slab layout. Reject the copy at the sink. The destination's INDEX_HDR already reports hdr_total (the payload capacity of the new buffer) and hdr_used (the bytes already consumed by the terminal END entry installed by indx_new()); require that to_move fits in the remaining payload before calling hdr_insert_head(). On mismatch, fail with -EINVAL and mark the filesystem as having a detected on-disk inconsistency, which is the same behaviour as the surrounding validation in this function. | ||||
| CVE-2026-72191 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: ntfs3: validate split-point offset in indx_insert_into_buffer indx_insert_into_buffer() computes used = used1 - to_copy - sp_size; memmove(de_t, Add2Ptr(sp, sp_size), used - le32_to_cpu(hdr1->de_off)); where sp and sp_size come from hdr_find_split(). hdr_find_split() walks entries by le16_to_cpu(e->size) without validating that each step stays within hdr->used or that the size field is at least sizeof(struct NTFS_DE). index_hdr_check(), the on-load gatekeeper, only validates header-level fields (used, total, de_off) and does not walk per-entry sizes. A crafted NTFS image whose leaf INDEX_HDR reports used == total but contains one interior NTFS_DE with size = 0xFFF0 therefore passes validation, descends to indx_insert_into_buffer() through the ntfs_create() -> indx_insert_entry() path, and makes hdr_find_split() return an sp whose sp_size (0xFFF0) greatly exceeds the remaining bytes in the buffer. The u32 subtraction underflows and the memmove count becomes a near-4-GiB value, producing an out-of-bounds kernel write that corrupts adjacent allocations and panics the kernel. Reproduced on 7.0.0-rc7 with UML + KASAN via a crafted image and a single 'touch' inside the mounted directory; crash site resolves to fs/ntfs3/index.c at the memmove. Trigger requires only local mount of an attacker-supplied filesystem image (USB, loopback, or removable media auto-mount). Reject the split whenever the chosen sp plus its declared size already extends past hdr1->used. This is the minimal fix; it preserves the existing hdr_find_split() contract and relies on the same out: cleanup path as the pre-existing error returns. A prior OOB read in the very same indx_insert_into_buffer() memmove was fixed in commit b8c44949044e ("fs/ntfs3: Fix OOB read in indx_insert_into_buffer") by tightening hdr_find_e(), but that fix does not cover the split-point size field path addressed here: sp is returned by hdr_find_split(), not hdr_find_e(), and the underflow is driven by sp->size rather than hdr->used exceeding hdr->total. | ||||
| CVE-2026-72190 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: ntfs: fix mrec_lock ABBA deadlock in rename ntfs_file_fsync(), ntfs_dir_fsync() and __ntfs_write_inode() lock an inode's mrec_lock before taking the mrec_lock of its parent directory. ntfs_rename() takes old_ni->mrec_lock and old_dir_ni->mrec_lock before taking new_ni->mrec_lock for an existing target, or new_dir_ni->mrec_lock for a cross-directory rename. This can deadlock when ntfs_file_fsync() or __ntfs_write_inode() holds the target inode, or when ntfs_dir_fsync() holds a child target directory, while rename() holds the parent directory and waits for the target. Fix this by locking the existing target inode before taking any parent directory mrec_lock. For cross-directory renames where the target parent is a descendant of the source parent, lock the target parent before the source parent so the directory order matches the child-to-parent order used by ntfs_file_fsync(), ntfs_dir_fsync(), and __ntfs_write_inode(). | ||||
| CVE-2026-72189 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: ntfs: fail attrlist updates when the superblock is inactive generic_shutdown_super() clears SB_ACTIVE before evicting cached inodes. If eviction selects the fake inode for a base inode's unnamed $ATTRIBUTE_LIST attribute, ntfs_evict_big_inode() drops the fake inode's reference on the base inode while the fake inode is still hashed and marked I_FREEING. That iput can synchronously write back the base inode. The writeback path may update mapping pairs and call ntfs_attrlist_update(), which unconditionally calls ntfs_attr_iget() for the same $ATTRIBUTE_LIST fake inode. VFS then finds the I_FREEING inode and waits for eviction to finish, but the current task is still inside that eviction path, causing a self-deadlock in find_inode(). Fix this by mirroring the teardown guard used by __ntfs_write_inode(): once SB_ACTIVE has been cleared, do not try to iget the attribute-list fake inode. Return -EIO so teardown aborts the update instead of waiting on the inode it is evicting. | ||||
| CVE-2026-72188 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: ntfs: sanitize MFT references returned from ntfs_lookup_inode_by_name() ntfs_lookup_inode_by_name() returns MFT references read from directory index entries on disk. These values are untrusted, but the function can currently return an error-marked MFT reference to its callers without validating it. Callers later decode lookup failures with MREF_ERR(). A crafted NTFS image can set the MREF error bit while leaving the low bits as an arbitrary value, causing callers to consume a bogus pseudo-errno instead of treating the lookup result as corrupted on-disk metadata. Fix this at the source by normalizing every error-marked MFT reference returned from ntfs_lookup_inode_by_name() to ERR_MREF(-EIO). Apply this to all four directory lookup return paths so every caller gets a validated result without needing additional checks or an API change. This keeps the sanitization in the common lookup helper, which is cleaner than duplicating validation in each caller. | ||||
| CVE-2026-72187 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: ntfs: avoid self-deadlock during inode eviction An attribute-list update performed while allocating clusters can drop the last reference to the temporary attribute inode. Evicting that inode drops its reference to the base inode and can invoke ntfs_drop_big_inode() for the base inode from within the base inode's own writeback path. If the base inode is unlinked, ntfs_drop_big_inode() calls truncate_setsize(), which waits for the inode's folio writeback to complete. The same writeback worker is responsible for completing that writeback, so it waits for itself indefinitely. Prevent this self-deadlock by grabbing a reference to the base inode at the beginning of ntfs_writepages() and releasing it at the end of the function. This defers eviction until all bios have been submitted, allowing the wait for folio writeback to complete safely. | ||||
| CVE-2026-72186 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: ntfs: make system files immutable to prevent corruption When a system file such as $Bitmap is exposed via show_sys_files and written from userspace, the volume is corrupted and, because the cluster allocator scans $Bitmap through the same inode's page cache, a write to $Bitmap also deadlocks writeback against the folio it already holds locked. These files are maintained by the driver itself and have no valid reason to be written through the file interface. Mark base metadata files (mft_no < FILE_first_user) as immutable during inode read so the VFS rejects write, mmap, truncate and unlink with -EPERM. Directories are skipped so the root and $Extend remain usable. Internal metadata updates do not go through the VFS write path and are unaffected. | ||||
| CVE-2026-72185 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: ntfs: fix WARN_ON for resident attribute in ntfs_map_runlist_nolock() When ntfs_map_runlist_nolock() needs to look up the attribute extent containing a target VCN (ctx_needs_reset == true), it calls ntfs_attr_lookup() and then expects the result to be a non-resident attribute, since only non-resident attributes have a mapping pairs array to decompress. A crafted NTFS image can place a resident attribute where a non-resident one is expected, causing ntfs_attr_lookup() to succeed but return a resident attribute record. Previously this was caught only by a WARN_ON(), which does not stop execution. The code then falls through to read a->data.non_resident.highest_vcn from what is actually a resident attribute, accessing the wrong union member and corrupting the VCN range check. The caller path triggering this warning during mount is: ntfs_map_runlist_nolock ntfs_empty_logfile load_system_files ntfs_fill_super In this path ctx is NULL, so ntfs_map_runlist_nolock() allocates a temporary search context internally and sets ctx_needs_reset = true. The existing resident-attribute guard in the ctx != NULL branch already returns -EIO silently for the same condition; make the ctx_needs_reset path consistent by replacing the WARN_ON() with the same -EIO error return. This causes the crafted image to be rejected with a mount error instead of triggering a kernel warning. | ||||
| CVE-2026-72184 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: ntfs: fix hole runlist memory leak in insert range error path ntfs_non_resident_attr_insert_range() allocates hole_rl before mapping the whole runlist. If ntfs_attr_map_whole_runlist() fails, the error path drops ni->runlist.lock and returns without freeing hole_rl. This leaks memory of sizeof(*hole_rl) * 2 bytes. Fix this memory leak by freeing hole_rl before returning from that error path, matching the later error paths in the same function. | ||||
| CVE-2026-72183 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: landlock: Fix LANDLOCK_SCOPE_SIGNAL bypass on the SIGIO path LANDLOCK_SCOPE_SIGNAL must prevent a sandboxed process from signaling processes outside its Landlock domain. It can be bypassed through the asynchronous SIGIO delivery path. A sandboxed process that owns any file or socket can arm it with fcntl(fd, F_SETOWN, -pgid), fcntl(fd, F_SETSIG, SIGKILL) and O_ASYNC, so that an I/O event makes the kernel deliver the chosen signal to the whole process group. As the head of its process group's task list (the default position right after fork()) that group can also hold the non-sandboxed process that launched it, e.g. a supervisor or a security monitor. The sandbox can thus kill or signal the processes LANDLOCK_SCOPE_SIGNAL is meant to protect from it. The scope is enforced in hook_file_send_sigiotask() against the Landlock domain recorded at F_SETOWN time, not the live domain of the sender. control_current_fowner() decides whether to record that domain and skips recording it when the fowner target is in the caller's thread group, which is safe only for a single-task target (PIDTYPE_PID, PIDTYPE_TGID). For a process group (PIDTYPE_PGID) pid_task() returns only one member; recording is skipped whenever that member shares the caller's thread group, and hook_file_send_sigiotask() then lets the signal fan out to the whole group unchecked. Record the domain for every non single-process target so the scope is enforced against each group member at delivery time. That recording is necessary but not sufficient on its own: the kernel signals a process group through its members' thread-group leaders, and the leader of the registrant's own process can carry a different Landlock domain than the sibling thread that armed the owner. domain_is_scoped() would then deny that leader, even though commit 18eb75f3af40 ("landlock: Always allow signals between threads of the same process") requires same-process delivery to be allowed. hook_task_kill() avoids this by evaluating same_thread_group() live, per recipient; the SIGIO path instead delegates the whole decision to a single registration-time check, which a process-group fan-out cannot honor. So also record the registrant's thread group next to its domain and exempt it at delivery: hook_file_send_sigiotask() allows the signal whenever the recipient belongs to the registrant's own process, restoring the same-process guarantee while keeping out-of-domain group members blocked. The direct kill() path (hook_task_kill) already evaluates the live domain and is unaffected. [mic: Check pid_type earlier and improve comment, fix commit message, fix comment formatting] | ||||
| CVE-2026-72182 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: power: supply: charger-manager: fix refcount leak in is_full_charged() In is_full_charged(), power_supply_get_by_name() is called to obtain a reference to the fuel_gauge power supply. If the voltage check (uV >= desc->fullbatt_uV) succeeds, the function returns true directly without releasing the reference, leaking the refcount. Fix this by setting a flag and jumping to the out label where power_supply_put() properly drops the reference. | ||||
| CVE-2026-72181 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: mips: sched: Fix CPUMASK_OFFSTACK memory corruption This patch addresses a critical memory management flaw. When CONFIG_CPUMASK_OFFSTACK is enabled, cpumask_var_t is a pointer. Consequently, sizeof(new_mask) evaluates to the pointer size, causing copy_from_user() to clobber the mask pointer. Furthermore, the old logic performed copy_from_user() before allocating the mask. Fix this by allocating new_mask first. To handle variable-sized user masks correctly, use cpumask_size() to truncate overly large user masks or pad undersized masks with zeros before copying the data directly into the allocated buffer. | ||||
| CVE-2026-72180 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: mm/huge_memory: preserve pmd_swp_uffd_wp on device-private PMD downgrade change_non_present_huge_pmd() rewrites a writable device-private PMD swap entry into a readable one without carrying pmd_swp_uffd_wp() across. The PTE-level change_softleaf_pte() does this correctly; mirror that here, matching what copy_huge_pmd() does for the fork path. Without the carry, a plain mprotect() over a UFFD_WP-marked device-private THP strips the bit and the trap is bypassed on swap-in. | ||||
| CVE-2026-72179 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: riscv: cacheinfo: Fix node reference leak in populate_cache_leaves Currently, the while loop drops the reference to prev in each iteration. If the loop terminates early due to a break, the final of_node_put(np) correctly drops the reference to the current node. However, if the loop terminates naturally because np == NULL, calling of_node_put(np) is a no-op. This leaves the last valid node stored in prev without its reference dropped, resulting in a node reference leak. Fix this by changing the final `of_node_put(np)` to `of_node_put(prev)`. | ||||
| CVE-2026-72178 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: mm/damon/core: always put unsuccessfully committed target pids damon_commit_target() puts and gets the destination and the source target pids. It puts the destination target pid because it will be overwritten by the source target pid. It gets the source pid because the caller is supposed to eventually put the pids. In more detail, the caller will call damon_destroy_ctx() after damon_commit_ctx() to destroy the entire source context. And in this case, [f]vaddr operation set's cleanup_target() callback will put the pids. The commit operation is made at the context level. The operation can fail in multiple places including in the middle and after the targets commit operations. For any such failures, immediately the error is returned to the damon_commit_ctx() caller. If some or all of the source target pids were committed to the destination during the unsuccessful context commit attempt, those pids should be put twice. The source context will do the put operations using the above explained routine. However, let's suppose the destination context was not originally using [f]vaddr operation set and the commit failed before the ops of the source context is committed. The destination does not have the cleanup_target() ops callback, so it cannot put the pids via the damon_destroy_ctx(). As a result, the pids are leaked. The issue in the real world would be not very common. The commit feature is for changing parameters of running DAMON context while inheriting internal status like the monitoring results. The monitoring results of a physical address range ain't have things that are beneficial to be inherited to a virtual address ranges monitoring. So the problem-causing DAMON control would be not very common in the real world. That said, it is a supported feature. And damon_commit_target() failure due to memory allocation is relatively realistic [1] if there are a huge number of target regions. Fix by putting the pids in the commit operation in case of the failures. The issue was discovered [2] by Sashiko. | ||||
| CVE-2026-72177 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: mm/damon/sysfs-schemes: fix dir put orders in access_pattern_add_dirs() Patch series "mm/damon/sysfs-schemes: fix wrong directories put orders in error paths". Error paths of damon_sysfs_access_pattern_add_dirs() and damon_sysfs_scheme_add_dirs() functions put references to directories in wrong orders. As a result, uninitialized memory dereference and/or memory leak can happen. Fix those. This patch (of 2): In access_pattern_add_dirs(), error handling path puts references starting from setup failed directories. If the failure happpened from the initial allication in the setup functions, uninitialized memory dereference happen. The allocation failures will not commonly happen, but the consequence is quite bad. Fix the wrong reference put orders. The issue was discovered [1] by Sashiko. | ||||
| CVE-2026-72176 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: mm/damon/sysfs-schemes: put stats for scheme_add_dirs() internal error damon_sysfs_scheme_add_dirs() setup the tried_regions directory after the stats directory setup is completed. When the tried_regions directory setup is failed, the setup function ensures the reference for the tried regions directory is released. Hence the error path should put references on setup succeeded directory objects, starting from the stats directory. However, the error path is putting the tried_regions directory instead of the stats directory. As a direct result, the stats directory object is leaked. Worse yet, if the tried_regions directory setup failed from the initial allocation, the scheme->tried_regions field remains uninitialized. The following kobject_put(&scheme->tried_regions->kobj) call in the error path will dereference the uninitialized memory. The setup failures should not be common. But once it happens, the consequence is quite bad. Fix this issue by correctly putting the stats directory instead of the tried_regions directory. The issue was discovered [1] by Sashiko. | ||||
| CVE-2026-72175 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: fs/proc/task_mmu: fix make_uffd_wp_huge_pte() prot-update race Patch series "userfaultfd/pagemap: pre-existing fixes". These are pre-existing bug fixes that were carried at the front of the userfaultfd RWP working-set-tracking series up to v5 [1]. Per review feedback that fixes should not sit in the middle of a feature series, they are split out and sent on their own; the RWP series is reposted rebased on top of this. All six were flagged by the Sashiko AI review of the RWP series and carry independent of RWP, apply to mm-new directly, and carry Cc: stable@. 1: fs/proc/task_mmu: a missing huge_ptep_modify_prot_start() in make_uffd_wp_huge_pte() can lose hardware Dirty/Accessed updates when PAGEMAP_SCAN write-protects a hugetlb PTE. 2: fs/proc/task_mmu: pagemap_scan_hugetlb_entry() compares the range against HPAGE_SIZE rather than the hstate page size, so it never write-protects gigantic hugetlb pages. 3: fs/proc/task_mmu: PAGEMAP_SCAN with PM_SCAN_WP_MATCHING over an unpopulated hugetlb range self-deadlocks -- pagemap_scan_pte_hole() calls uffd_wp_range() while walk_hugetlb_range() holds the hugetlb vma lock for read, and hugetlb_change_protection() then takes it for write. Install the marker inline instead. 4: mm/huge_memory: change_non_present_huge_pmd() drops pmd_swp_uffd_wp on a device-private PMD permission downgrade, silently losing the uffd-wp marker. 5: userfaultfd: must_wait() applies pte_write() to a locklessly read PTE without checking pte_present(), so swap/migration entries decode random offset bits and a thread can stay parked on a stale fault. 6: userfaultfd: __VMA_UFFD_FLAGS feeds VMA_UFFD_MINOR_BIT (41) to mk_vma_flags() unconditionally, an out-of-bounds write into the single-word vma_flags_t on 32-bit. Build the mask from config-gated per-mode masks so an unavailable bit is never materialised. This patch (of 6): make_uffd_wp_huge_pte() arms the UFFD_WP bit on a present HugeTLB PTE by calling huge_ptep_modify_prot_commit() with a ptent snapshot that was fetched without the corresponding huge_ptep_modify_prot_start(). The start helper is what atomically clears the entry so the kernel-owned snapshot stays consistent until the commit; without it, the hardware may set Dirty or Accessed in the live PTE between the original read and the commit, and huge_ptep_modify_prot_commit() (whose generic implementation just calls set_huge_pte_at()) then writes the stale snapshot back over the live hardware bits, losing the update. The non-hugetlb sibling make_uffd_wp_pte() does this correctly via ptep_modify_prot_start() / ptep_modify_prot_commit(). Mirror that pattern for the present-PTE branch. The migration case stays as-is -- migration entries are non-present, so there's no hardware update to race against. | ||||
| CVE-2026-72174 | 1 Linux | 1 Linux Kernel | 2026-08-15 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: fs/proc/task_mmu: fix hugetlb self-deadlock in pagemap_scan_pte_hole() A PAGEMAP_SCAN ioctl requesting PM_SCAN_WP_MATCHING on a hugetlb VMA hangs the calling thread, unkillably, as soon as the scan reaches an unpopulated part of the range: do_pagemap_scan() walk_page_range() walk_hugetlb_range() hugetlb_vma_lock_read() # take the vma lock for read ... pagemap_scan_pte_hole() # ... ->pte_hole() for a hole uffd_wp_range() change_protection() hugetlb_change_protection() hugetlb_vma_lock_write() # ... and block taking it for write walk_hugetlb_range() holds the hugetlb vma lock for read across the whole walk. A present entry goes to ->hugetlb_entry(); an unpopulated one goes to ->pte_hole(), i.e. pagemap_scan_pte_hole(). To write-protect the hole that handler calls uffd_wp_range(), which on a hugetlb VMA reaches hugetlb_change_protection() and takes the same vma lock for write. The thread then blocks in down_write() waiting for the read lock it is itself holding. The populated path avoids this: pagemap_scan_hugetlb_entry() write-protects the entry inline under the page-table lock and never enters hugetlb_change_protection(). Do the same for holes. Fault in the page table and install the uffd-wp marker directly with make_uffd_wp_huge_pte() under the page-table lock, rather than routing through uffd_wp_range(). That is the same sequence hugetlb_change_protection() runs for an unpopulated entry, minus the vma write lock -- which is safe to skip because PMD sharing is disabled on uffd-wp VMAs (hugetlb_unshare_all_pmds() runs at registration), leaving nothing for that lock to serialise against. | ||||