From 1039bffd6ae9c75b42b7d148d6c1106134107b66 Mon Sep 17 00:00:00 2001 From: Xie Yuanbin Date: Tue, 28 Jul 2026 03:16:42 +0100 Subject: ARM: 9485/1: mm: acquire mmap write lock around show_pte() for user faults When CONFIG_DEBUG_USER=y, and cmdline "user_debug=31" is set, a user fault may trigger show_pte() without any lock. If another thread in the same process concurrently calls munmap(), the page table pages may be freed while show_pte() is still traversing them, causing a use-after-free in show_pte(). If CONFIG_ARM_LPAE=y, this may cause a kernel panic if the pages table of PMD are freed when show_pte() is running. Acquire mmap_write_lock() around show_pte() for user faults to fix the contention. For user faults, additionally restrict that show_pte() is called only when the addr is a user-space address (addr < TASK_SIZE). This is because the lock of tsk->mm only protects the virtual memory of user address space, furthermore, dumping the page tables of a kernel-space address for user faults is unnecessary and may have security implications. Keep everything unchanged for kernel faults, because the kernel is already in the "oops" state, acquiring a lock may risk a deadlock. Co-developed-by: Qi Xi Fixes: 6d021b724481 ("ARM: dump pgd, pmd and pte states on unhandled data abort faults") Link: https://lore.kernel.org/20260716014022.2823-1-xieyuanbin1@huawei.com Acked-by: Lorenzo Stoakes (ARM) Reviewed-by: Linus Walleij Signed-off-by: Qi Xi Signed-off-by: Xie Yuanbin Signed-off-by: Russell King --- tools/testing/vma/shared.c | 131 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 131 insertions(+) create mode 100644 tools/testing/vma/shared.c (limited to 'tools/testing/vma/shared.c') diff --git a/tools/testing/vma/shared.c b/tools/testing/vma/shared.c new file mode 100644 index 000000000..2565a5aec --- /dev/null +++ b/tools/testing/vma/shared.c @@ -0,0 +1,131 @@ +// SPDX-License-Identifier: GPL-2.0-or-later + +#include "shared.h" + + +bool fail_prealloc; +unsigned long mmap_min_addr = CONFIG_DEFAULT_MMAP_MIN_ADDR; +unsigned long dac_mmap_min_addr = CONFIG_DEFAULT_MMAP_MIN_ADDR; +unsigned long stack_guard_gap = 256UL<vm_start = start; + vma->vm_end = end; + vma->vm_pgoff = pgoff; + vma->flags = vma_flags; + vma_assert_detached(vma); + + return vma; +} + +void detach_free_vma(struct vm_area_struct *vma) +{ + vma_mark_detached(vma); + vm_area_free(vma); +} + +struct vm_area_struct *alloc_and_link_vma(struct mm_struct *mm, + unsigned long start, unsigned long end, + pgoff_t pgoff, vma_flags_t vma_flags) +{ + struct vm_area_struct *vma = alloc_vma(mm, start, end, pgoff, vma_flags); + + if (vma == NULL) + return NULL; + + if (attach_vma(mm, vma)) { + detach_free_vma(vma); + return NULL; + } + + /* + * Reset this counter which we use to track whether writes have + * begun. Linking to the tree will have caused this to be incremented, + * which means we will get a false positive otherwise. + */ + vma->vm_lock_seq = UINT_MAX; + + return vma; +} + +void reset_dummy_anon_vma(void) +{ + dummy_anon_vma.was_cloned = false; + dummy_anon_vma.was_unlinked = false; +} + +int cleanup_mm(struct mm_struct *mm, struct vma_iterator *vmi) +{ + struct vm_area_struct *vma; + int count = 0; + + fail_prealloc = false; + reset_dummy_anon_vma(); + + vma_iter_set(vmi, 0); + for_each_vma(*vmi, vma) { + detach_free_vma(vma); + count++; + } + + mtree_destroy(&mm->mm_mt); + mm->map_count = 0; + return count; +} + +bool vma_write_started(struct vm_area_struct *vma) +{ + int seq = vma->vm_lock_seq; + + /* We reset after each check. */ + vma->vm_lock_seq = UINT_MAX; + + /* The vma_start_write() stub simply increments this value. */ + return seq > -1; +} + +void __vma_set_dummy_anon_vma(struct vm_area_struct *vma, + struct anon_vma_chain *avc, struct anon_vma *anon_vma) +{ + vma->anon_vma = anon_vma; + INIT_LIST_HEAD(&vma->anon_vma_chain); + list_add(&avc->same_vma, &vma->anon_vma_chain); + avc->anon_vma = vma->anon_vma; +} + +void vma_set_dummy_anon_vma(struct vm_area_struct *vma, + struct anon_vma_chain *avc) +{ + __vma_set_dummy_anon_vma(vma, avc, &dummy_anon_vma); +} + +struct task_struct *get_current(void) +{ + return &__current; +} + +unsigned long rlimit(unsigned int limit) +{ + return (unsigned long)-1; +} + +void vma_set_range(struct vm_area_struct *vma, + unsigned long start, unsigned long end, + pgoff_t pgoff) +{ + vma->vm_start = start; + vma->vm_end = end; + vma->vm_pgoff = pgoff; +} -- cgit v1.3.1