From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751788AbeEQMrI (ORCPT ); Thu, 17 May 2018 08:47:08 -0400 Received: from mail-pl0-f68.google.com ([209.85.160.68]:42916 "EHLO mail-pl0-f68.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750924AbeEQMrH (ORCPT ); Thu, 17 May 2018 08:47:07 -0400 X-Google-Smtp-Source: AB8JxZq8cjUp26MYSlMI2JufxoQsYqjV5mb2z0MyBhJdNV4xwWU5dFyuSpUpCoi8D8VEvItDAHGyyQ== Subject: Re: [PATCH] KVM: arm/arm64: add WARN_ON if size is not PAGE_SIZE aligned in unmap_stage2_range To: Suzuki K Poulose , Christoffer Dall , Marc Zyngier , linux-arm-kernel@lists.infradead.org, kvmarm@lists.cs.columbia.edu Cc: Andrew Morton , Andrea Arcangeli , Claudio Imbrenda , Arvind Yadav , "David S. Miller" , Minchan Kim , Mike Rapoport , Hugh Dickins , "Paul E. McKenney" , linux-mm@kvack.org, linux-kernel@vger.kernel.org, jia.he@hxt-semitech.com References: <1526537487-14804-1-git-send-email-hejianet@gmail.com> <698b0355-d430-86b8-cd09-83c6d9e566f8@arm.com> From: Jia He Message-ID: Date: Thu, 17 May 2018 20:46:50 +0800 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:52.0) Gecko/20100101 Thunderbird/52.7.0 MIME-Version: 1.0 In-Reply-To: <698b0355-d430-86b8-cd09-83c6d9e566f8@arm.com> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Suzuki On 5/17/2018 4:17 PM, Suzuki K Poulose Wrote: > > Hi Jia, > > On 17/05/18 07:11, Jia He wrote: >> I ever met a panic under memory pressure tests(start 20 guests and run >> memhog in the host). > > Please avoid using "I" in the commit description and preferably stick to > an objective description. Thanks for the pointing > >> >> The root cause might be what I fixed at [1]. But from arm kvm points of >> view, it would be better we caught the exception earlier and clearer. >> >> If the size is not PAGE_SIZE aligned, unmap_stage2_range might unmap the >> wrong(more or less) page range. Hence it caused the "BUG: Bad page >> state" > > I don't see why we should ever panic with a "positive" size value. Anyways, > the unmap requests must be in units of pages. So this check might be useful. > > good question, After further digging, maybe we need to harden the break condition as below? diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c index 7f6a944..dac9b2e 100644 --- a/virt/kvm/arm/mmu.c +++ b/virt/kvm/arm/mmu.c @@ -217,7 +217,7 @@ static void unmap_stage2_ptes(struct kvm *kvm, pmd_t *pmd, put_page(virt_to_page(pte)); } - } while (pte++, addr += PAGE_SIZE, addr != end); + } while (pte++, addr += PAGE_SIZE, addr < end); basically verified in my armv8a server -- Cheers, Jia > Reviewed-by: Suzuki K Poulose > >> >> [1] https://lkml.org/lkml/2018/5/3/1042 >> >> Signed-off-by: jia.he@hxt-semitech.com >> --- >>   virt/kvm/arm/mmu.c | 2 ++ >>   1 file changed, 2 insertions(+) >> >> diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c >> index 7f6a944..8dac311 100644 >> --- a/virt/kvm/arm/mmu.c >> +++ b/virt/kvm/arm/mmu.c >> @@ -297,6 +297,8 @@ static void unmap_stage2_range(struct kvm *kvm, >> phys_addr_t start, u64 size) >>       phys_addr_t next; >>         assert_spin_locked(&kvm->mmu_lock); >> +    WARN_ON(size & ~PAGE_MASK); >> + >>       pgd = kvm->arch.pgd + stage2_pgd_index(addr); >>       do { >>           /* >> > >