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 0C277C433FE for ; Wed, 16 Nov 2022 15:53:19 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233101AbiKPPxQ (ORCPT ); Wed, 16 Nov 2022 10:53:16 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34568 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230287AbiKPPxN (ORCPT ); Wed, 16 Nov 2022 10:53:13 -0500 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id 272CA51C18 for ; Wed, 16 Nov 2022 07:53:11 -0800 (PST) Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id A22691477; Wed, 16 Nov 2022 07:53:13 -0800 (PST) Received: from FVFF77S0Q05N.cambridge.arm.com (FVFF77S0Q05N.cambridge.arm.com [10.1.35.169]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 3FCD53F587; Wed, 16 Nov 2022 07:53:02 -0800 (PST) Date: Wed, 16 Nov 2022 15:52:58 +0000 From: Mark Rutland To: Liu Shixin Cc: Catalin Marinas , Will Deacon , Denys Vlasenko , Kefeng Wang , Anshuman Khandual , David Hildenbrand , Rafael Aquini , Pasha Tatashin , linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 2/2] arm64/mm: fix incorrect file_map_count for invalid pmd/pud Message-ID: References: <20221116083811.464678-1-liushixin2@huawei.com> <20221116083811.464678-3-liushixin2@huawei.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20221116083811.464678-3-liushixin2@huawei.com> Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Nov 16, 2022 at 04:38:11PM +0800, Liu Shixin wrote: > The page table check trigger BUG_ON() unexpectedly when split hugepage: > > ------------[ cut here ]------------ > kernel BUG at mm/page_table_check.c:119! > Internal error: Oops - BUG: 00000000f2000800 [#1] SMP > Dumping ftrace buffer: > (ftrace buffer empty) > Modules linked in: > CPU: 7 PID: 210 Comm: transhuge-stres Not tainted 6.1.0-rc3+ #748 > Hardware name: linux,dummy-virt (DT) > pstate: 20000005 (nzCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) > pc : page_table_check_set.isra.0+0x398/0x468 > lr : page_table_check_set.isra.0+0x1c0/0x468 > [...] > Call trace: > page_table_check_set.isra.0+0x398/0x468 > __page_table_check_pte_set+0x160/0x1c0 > __split_huge_pmd_locked+0x900/0x1648 > __split_huge_pmd+0x28c/0x3b8 > unmap_page_range+0x428/0x858 > unmap_single_vma+0xf4/0x1c8 > zap_page_range+0x2b0/0x410 > madvise_vma_behavior+0xc44/0xe78 > do_madvise+0x280/0x698 > __arm64_sys_madvise+0x90/0xe8 > invoke_syscall.constprop.0+0xdc/0x1d8 > do_el0_svc+0xf4/0x3f8 > el0_svc+0x58/0x120 > el0t_64_sync_handler+0xb8/0xc0 > el0t_64_sync+0x19c/0x1a0 > [...] > > On arm64, pmd_present() will return true even if the pmd is invalid. So > in pmdp_invalidate() the file_map_count will not only decrease once but > also increase once. Then in set_pte_at(), the file_map_count increase > again, and so trigger BUG_ON() unexpectedly. It's not clear to me how pmd_present() relates to p?d_user_accessible_page() below. How are they related? (or is this a copy-paste error)? > Fix this problem by adding pmd_valid() in pmd_user_accessible_page(). > Moreover, add pud_valid() for pud_user_accessible_page() too. > > Fixes: 42b2547137f5 ("arm64/mm: enable ARCH_SUPPORTS_PAGE_TABLE_CHECK") > Reported-by: Denys Vlasenko > Signed-off-by: Liu Shixin > --- > arch/arm64/include/asm/pgtable.h | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h > index edf6625ce965..56e178de75e7 100644 > --- a/arch/arm64/include/asm/pgtable.h > +++ b/arch/arm64/include/asm/pgtable.h > @@ -863,12 +863,12 @@ static inline bool pte_user_accessible_page(pte_t pte) > > static inline bool pmd_user_accessible_page(pmd_t pmd) > { > - return pmd_leaf(pmd) && (pmd_user(pmd) || pmd_user_exec(pmd)); > + return pmd_leaf(pmd) && (pmd_user(pmd) || pmd_user_exec(pmd)) && pmd_valid(pmd); > } > > static inline bool pud_user_accessible_page(pud_t pud) > { > - return pud_leaf(pud) && pud_user(pud); > + return pud_leaf(pud) && pud_user(pud) && pud_valid(pud); I think these p?d_valid() checks should be first for clarity, since the other bits aren't necessarily meaningful for !valid entries. Thanks, Mark.