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 X-Spam-Level: X-Spam-Status: No, score=-4.4 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI, SPF_HELO_NONE,SPF_PASS autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 9D68BC433ED for ; Tue, 4 May 2021 17:42:31 +0000 (UTC) Received: from desiato.infradead.org (desiato.infradead.org [90.155.92.199]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 200B7613C6 for ; Tue, 4 May 2021 17:42:31 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 200B7613C6 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=arm.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-arm-kernel-bounces+linux-arm-kernel=archiver.kernel.org@lists.infradead.org DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=desiato.20200630; h=Sender:Content-Transfer-Encoding :Content-Type:List-Subscribe:List-Help:List-Post:List-Archive: List-Unsubscribe:List-Id:In-Reply-To:MIME-Version:References:Message-ID: Subject:Cc:To:From:Date:Reply-To:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Owner; bh=1/ryxtv7xJ2iAeIiayts2UnLMntQ7gVgw3Hbnn1OYBE=; b=Lb9e5nZCsjFhWxGm3ogXcYsZx arXBX2hsK04Ou9bD8qPN3Vesu6wwn5Lbhxfn38RdhfKxuM6+9iKb+3EDwwnxtHVT8fnDFzLEvSWQv mLKX4uexmsFbQ1Sk+NAfjLvlywXeBhLGVdg8DSxrk0Ux9ce3LkbV+YGwI9GEUGbxvHlM5XgiaI191 GgCf295WqmRnRRlDnRky3LXSjupDaTF8gCyonD1vbUYiSWkvPozHhk8qHvx+F4uQoPy9Y2C5OZ6QU M/T+8N8eAOIxr6LgwS4I22uG6rIfhXb6ZCSxKlOORgiK26qPPzEqwoJmH/XLHmr+QP7Ue4C39uNbk Pg7Z3EdCw==; Received: from localhost ([::1] helo=desiato.infradead.org) by desiato.infradead.org with esmtp (Exim 4.94 #2 (Red Hat Linux)) id 1ldz26-00GhII-Af; Tue, 04 May 2021 17:40:30 +0000 Received: from bombadil.infradead.org ([2607:7c80:54:e::133]) by desiato.infradead.org with esmtps (Exim 4.94 #2 (Red Hat Linux)) id 1ldz24-00GhHu-CN for linux-arm-kernel@desiato.infradead.org; Tue, 04 May 2021 17:40:28 +0000 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=infradead.org; s=bombadil.20210309; h=In-Reply-To:Content-Type:MIME-Version :References:Message-ID:Subject:Cc:To:From:Date:Sender:Reply-To: Content-Transfer-Encoding:Content-ID:Content-Description; bh=dFP3ol/2yMSiy8MEye34cgcrOS06DGfGX/FDzNf0heI=; b=OYGqwQE/vNxAUxaT7sU7YwsAVc 4sw5BillqGcKORT0ij0KiReeefU00gBZ1lGIDiVOU4iZ3QCphzYMYsPSMd7p48CQLPSifJ0fDvDc5 doBtYz+lpb37F/Ay3l5WsRFCT2MpnNvMqXFXzFE3Dz3PFP6aymSI/htY8IO4Neco1Uw8GgnLJW1RF z0wFZm2Ng7ooUvfGn6lfTM/jfpNKxOQrY2TDP78Pw5iBRVM/ReOSuTYgokUGrP690VUA7S957EuUW 1LWr/5LLX2mzuAXZFHv0ATuiZoc4KN1JBATUI5TAdqZHbKHh/NDc3xg1uD5nqf+RxXiODcw7esXR/ VfVZHGvA==; Received: from mail.kernel.org ([198.145.29.99]) by bombadil.infradead.org with esmtps (Exim 4.94 #2 (Red Hat Linux)) id 1ldz21-0049Cg-89 for linux-arm-kernel@lists.infradead.org; Tue, 04 May 2021 17:40:27 +0000 Received: by mail.kernel.org (Postfix) with ESMTPSA id 3B4F6613CB; Tue, 4 May 2021 17:40:21 +0000 (UTC) Date: Tue, 4 May 2021 18:40:18 +0100 From: Catalin Marinas To: Steven Price Cc: Marc Zyngier , Will Deacon , James Morse , Julien Thierry , Suzuki K Poulose , kvmarm@lists.cs.columbia.edu, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, Dave Martin , Mark Rutland , Thomas Gleixner , qemu-devel@nongnu.org, Juan Quintela , "Dr. David Alan Gilbert" , Richard Henderson , Peter Maydell , Haibo Xu , Andrew Jones Subject: Re: [PATCH v11 2/6] arm64: kvm: Introduce MTE VM feature Message-ID: References: <20210416154309.22129-1-steven.price@arm.com> <20210416154309.22129-3-steven.price@arm.com> <20210428170705.GB4022@arm.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20210504_104025_361415_786AE4FA X-CRM114-Status: GOOD ( 48.27 ) X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=archiver.kernel.org@lists.infradead.org On Thu, Apr 29, 2021 at 05:06:41PM +0100, Steven Price wrote: > On 28/04/2021 18:07, Catalin Marinas wrote: > > I probably asked already but is the only way to map a standard RAM page > > (not device) in stage 2 via the fault handler? One case I had in mind > > was something like get_user_pages() but it looks like that one doesn't > > call set_pte_at_notify(). There are a few other places where > > set_pte_at_notify() is called and these may happen before we got a > > chance to fault on stage 2, effectively populating the entry (IIUC). If > > that's an issue, we could move the above loop and check closer to the > > actual pte setting like kvm_pgtable_stage2_map(). > > The only call sites of kvm_pgtable_stage2_map() are in mmu.c: > > * kvm_phys_addr_ioremap() - maps as device in stage 2 > > * user_mem_abort() - handled above > > * kvm_set_spte_handler() - ultimately called from the .change_pte() > callback of the MMU notifier > > So the last one is potentially a problem. It's called via the MMU notifiers > in the case of set_pte_at_notify(). The users of that are: > > * uprobe_write_opcode(): Allocates a new page and performs a > copy_highpage() to copy the data to the new page (which with MTE includes > the tags and will copy across the PG_mte_tagged flag). > > * write_protect_page() (KSM): Changes the permissions on the PTE but it's > still the same page, so nothing to do regarding MTE. My concern here is that the VMM had a stage 1 pte but we haven't yet faulted in at stage 2 via user_mem_abort(), so we don't have any stage 2 pte set. write_protect_page() comes in and sets the new stage 2 pte via the callback. I couldn't find any check in kvm_pgtable_stage2_map() for the old pte, so it will set the new stage 2 pte regardless. A subsequent guest read would no longer fault at stage 2. > * replace_page() (KSM): If the page has MTE tags then the MTE version of > memcmp_pages() will return false, so the only caller > (try_to_merge_one_page()) will never call this on a page with tags. > > * wp_page_copy(): This one is more interesting - if we go down the > cow_user_page() path with an old page then everything is safe (tags are > copied over). The is_zero_pfn() case worries me a bit - a new page is > allocated, but I can't instantly see anything to zero out the tags (and set > PG_mte_tagged). True, I think tag zeroing happens only if we map it as PROT_MTE in the VMM. > * migrate_vma_insert_page(): I think migration should be safe as the tags > should be copied. > > So wp_page_copy() looks suspicious. > > kvm_pgtable_stage2_map() looks like it could be a good place for the checks, > it looks like it should work and is probably a more obvious place for the > checks. That would be my preference. It also matches the stage 1 set_pte_at(). > > While the set_pte_at() race on the page flags is somewhat clearer, we > > may still have a race here with the VMM's set_pte_at() if the page is > > mapped as tagged. KVM has its own mmu_lock but it wouldn't be held when > > handling the VMM page tables (well, not always, see below). > > > > gfn_to_pfn_prot() ends up calling get_user_pages*(). At least the slow > > path (hva_to_pfn_slow()) ends up with FOLL_TOUCH in gup and the VMM pte > > would be set, tags cleared (if PROT_MTE) before the stage 2 pte. I'm not > > sure whether get_user_page_fast_only() does the same. > > > > The race with an mprotect(PROT_MTE) in the VMM is fine I think as the > > KVM mmu notifier is invoked before set_pte_at() and racing with another > > user_mem_abort() is serialised by the KVM mmu_lock. The subsequent > > set_pte_at() would see the PG_mte_tagged set either by the current CPU > > or by the one it was racing with. > > Given the changes to set_pte_at() which means that tags are restored from > swap even if !PROT_MTE, the only race I can see remaining is the creation of > new PROT_MTE mappings. As you mention an attempt to change mappings in the > VMM memory space should involve a mmu notifier call which I think serialises > this. So the remaining issue is doing this in a separate address space. > > So I guess the potential problem is: > > * allocate memory MAP_SHARED but !PROT_MTE > * fork() > * VM causes a fault in parent address space > * child does a mprotect(PROT_MTE) > > With the last two potentially racing. Sadly I can't see a good way of > handling that. Ah, the mmap lock doesn't help as they are different processes (mprotect() acquires it as a writer). I wonder whether this is racy even in the absence of KVM. If both parent and child do an mprotect(PROT_MTE), one of them may be reading stale tags for a brief period. Maybe we should revisit whether shared MTE pages are of any use, though it's an ABI change (not bad if no-one is relying on this). However... Thinking about this, we have a similar problem with the PG_dcache_clean and two processes doing mprotect(PROT_EXEC). One of them could see the flag set and skip the I-cache maintenance while the other executes stale instructions. change_pte_range() could acquire the page lock if the page is VM_SHARED (my preferred core mm fix). It doesn't immediately solve the MTE/KVM case but we could at least take the page lock via user_mem_abort(). Or maybe we just document this behaviour as racy both for PROT_EXEC and PROT_MTE mappings and be done with this. The minor issue with PROT_MTE is the potential leaking of tags (it's harder to leak information through the I-cache). -- Catalin _______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-arm-kernel