From: Liu Shixin liushixin2@huawei.com
commit 958f32ce832ba781ac20e11bb2d12a9352ea28fc upstream.
The vma_lock and hugetlb_fault_mutex are dropped before handling userfault and reacquire them again after handle_userfault(), but reacquire the vma_lock could lead to UAF[1,2] due to the following race,
hugetlb_fault hugetlb_no_page /*unlock vma_lock */ hugetlb_handle_userfault handle_userfault /* unlock mm->mmap_lock*/ vm_mmap_pgoff do_mmap mmap_region munmap_vma_range /* clean old vma */ /* lock vma_lock again <--- UAF */ /* unlock vma_lock */
Since the vma_lock will unlock immediately after hugetlb_handle_userfault(), let's drop the unneeded lock and unlock in hugetlb_handle_userfault() to fix the issue.
[1] https://lore.kernel.org/linux-mm/000000000000d5e00a05e834962e@google.com/ [2] https://lore.kernel.org/linux-mm/20220921014457.1668-1-liuzixian4@huawei.com... Link: https://lkml.kernel.org/r/20220923042113.137273-1-liushixin2@huawei.com Fixes: 1a1aad8a9b7b ("userfaultfd: hugetlbfs: add userfaultfd hugetlb hook") Signed-off-by: Liu Shixin liushixin2@huawei.com Signed-off-by: Kefeng Wang wangkefeng.wang@huawei.com Reported-by: syzbot+193f9cee8638750b23cf@syzkaller.appspotmail.com Reported-by: Liu Zixian liuzixian4@huawei.com Reviewed-by: Mike Kravetz mike.kravetz@oracle.com Cc: David Hildenbrand david@redhat.com Cc: John Hubbard jhubbard@nvidia.com Cc: Muchun Song songmuchun@bytedance.com Cc: Sidhartha Kumar sidhartha.kumar@oracle.com Cc: stable@vger.kernel.org [4.14+] Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- mm/hugetlb.c | 29 +++++++++++++++-------------- 1 file changed, 15 insertions(+), 14 deletions(-)
--- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -4337,6 +4337,7 @@ static vm_fault_t hugetlb_no_page(struct spinlock_t *ptl; unsigned long haddr = address & huge_page_mask(h); bool new_page = false; + u32 hash = hugetlb_fault_mutex_hash(mapping, idx);
/* * Currently, we are forced to kill the process in the event the @@ -4346,7 +4347,7 @@ static vm_fault_t hugetlb_no_page(struct if (is_vma_resv_set(vma, HPAGE_RESV_UNMAPPED)) { pr_warn_ratelimited("PID %d killed due to inadequate hugepage pool\n", current->pid); - return ret; + goto out; }
/* @@ -4365,7 +4366,6 @@ retry: * Check for page in userfault range */ if (userfaultfd_missing(vma)) { - u32 hash; struct vm_fault vmf = { .vma = vma, .address = haddr, @@ -4380,17 +4380,14 @@ retry: };
/* - * hugetlb_fault_mutex and i_mmap_rwsem must be - * dropped before handling userfault. Reacquire - * after handling fault to make calling code simpler. + * vma_lock and hugetlb_fault_mutex must be dropped + * before handling userfault. Also mmap_lock will + * be dropped during handling userfault, any vma + * operation should be careful from here. */ - hash = hugetlb_fault_mutex_hash(mapping, idx); mutex_unlock(&hugetlb_fault_mutex_table[hash]); i_mmap_unlock_read(mapping); - ret = handle_userfault(&vmf, VM_UFFD_MISSING); - i_mmap_lock_read(mapping); - mutex_lock(&hugetlb_fault_mutex_table[hash]); - goto out; + return handle_userfault(&vmf, VM_UFFD_MISSING); }
page = alloc_huge_page(vma, haddr, 0); @@ -4497,6 +4494,8 @@ retry:
unlock_page(page); out: + mutex_unlock(&hugetlb_fault_mutex_table[hash]); + i_mmap_unlock_read(mapping); return ret;
backout: @@ -4592,10 +4591,12 @@ vm_fault_t hugetlb_fault(struct mm_struc mutex_lock(&hugetlb_fault_mutex_table[hash]);
entry = huge_ptep_get(ptep); - if (huge_pte_none(entry)) { - ret = hugetlb_no_page(mm, vma, mapping, idx, address, ptep, flags); - goto out_mutex; - } + if (huge_pte_none(entry)) + /* + * hugetlb_no_page will drop vma lock and hugetlb fault + * mutex internally, which make us return immediately. + */ + return hugetlb_no_page(mm, vma, mapping, idx, address, ptep, flags);
ret = 0;