From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 5A895FA3740 for ; Mon, 24 Oct 2022 14:17:44 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231585AbiJXORl (ORCPT ); Mon, 24 Oct 2022 10:17:41 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:58620 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S237155AbiJXOQB (ORCPT ); Mon, 24 Oct 2022 10:16:01 -0400 Received: from dfw.source.kernel.org (dfw.source.kernel.org [139.178.84.217]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 479F6B1CD; Mon, 24 Oct 2022 05:55:41 -0700 (PDT) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by dfw.source.kernel.org (Postfix) with ESMTPS id 614E061299; Mon, 24 Oct 2022 12:55:41 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 4574FC433D6; Mon, 24 Oct 2022 12:55:40 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1666616140; bh=CJQzFCSpz+lYJ0uCuw2di8h4ZYzJroJUUsO+QGDVdeE=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=sIMTjIyAI/u+WIcSjoroJVTh4LOPCTnnMb1LZSfLXnDUt6y8a8QPLhITncsMUqBLd IxZAX28JcDfTfwrIFdBXffdv8iRzicpMamyQrTus1bPqv47bikEnaKoeNANX9mRO+z XEkwl6FNsYgU1qtP3ngKyWS0RxHVly0npUHjC98Y= From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman , stable@vger.kernel.org, Liu Shixin , Kefeng Wang , syzbot+193f9cee8638750b23cf@syzkaller.appspotmail.com, Liu Zixian , Mike Kravetz , David Hildenbrand , John Hubbard , Muchun Song , Sidhartha Kumar , Andrew Morton Subject: [PATCH 5.15 516/530] mm: hugetlb: fix UAF in hugetlb_handle_userfault Date: Mon, 24 Oct 2022 13:34:20 +0200 Message-Id: <20221024113108.359729536@linuxfoundation.org> X-Mailer: git-send-email 2.38.1 In-Reply-To: <20221024113044.976326639@linuxfoundation.org> References: <20221024113044.976326639@linuxfoundation.org> User-Agent: quilt/0.67 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: stable@vger.kernel.org From: Liu Shixin 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 Signed-off-by: Kefeng Wang Reported-by: syzbot+193f9cee8638750b23cf@syzkaller.appspotmail.com Reported-by: Liu Zixian Reviewed-by: Mike Kravetz Cc: David Hildenbrand Cc: John Hubbard Cc: Muchun Song Cc: Sidhartha Kumar Cc: [4.14+] Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- mm/hugetlb.c | 37 +++++++++++++++++-------------------- 1 file changed, 17 insertions(+), 20 deletions(-) --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -4844,7 +4844,6 @@ static inline vm_fault_t hugetlb_handle_ unsigned long haddr, unsigned long reason) { - vm_fault_t ret; u32 hash; struct vm_fault vmf = { .vma = vma, @@ -4861,18 +4860,14 @@ static inline vm_fault_t hugetlb_handle_ }; /* - * 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, reason); - i_mmap_lock_read(mapping); - mutex_lock(&hugetlb_fault_mutex_table[hash]); - - return ret; + return handle_userfault(&vmf, reason); } static vm_fault_t hugetlb_no_page(struct mm_struct *mm, @@ -4889,6 +4884,7 @@ static vm_fault_t hugetlb_no_page(struct spinlock_t *ptl; unsigned long haddr = address & huge_page_mask(h); bool new_page, new_pagecache_page = false; + u32 hash = hugetlb_fault_mutex_hash(mapping, idx); /* * Currently, we are forced to kill the process in the event the @@ -4898,7 +4894,7 @@ 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; } /* @@ -4915,12 +4911,10 @@ retry: page = find_lock_page(mapping, idx); if (!page) { /* Check for page in userfault range */ - if (userfaultfd_missing(vma)) { - ret = hugetlb_handle_userfault(vma, mapping, idx, + if (userfaultfd_missing(vma)) + return hugetlb_handle_userfault(vma, mapping, idx, flags, haddr, VM_UFFD_MISSING); - goto out; - } page = alloc_huge_page(vma, haddr, 0); if (IS_ERR(page)) { @@ -4980,10 +4974,9 @@ retry: if (userfaultfd_minor(vma)) { unlock_page(page); put_page(page); - ret = hugetlb_handle_userfault(vma, mapping, idx, + return hugetlb_handle_userfault(vma, mapping, idx, flags, haddr, VM_UFFD_MINOR); - goto out; } } @@ -5034,6 +5027,8 @@ retry: unlock_page(page); out: + mutex_unlock(&hugetlb_fault_mutex_table[hash]); + i_mmap_unlock_read(mapping); return ret; backout: @@ -5131,10 +5126,12 @@ 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;