Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751183AbdH1Hpc convert rfc822-to-8bit (ORCPT ); Mon, 28 Aug 2017 03:45:32 -0400 Received: from tyo162.gate.nec.co.jp ([114.179.232.162]:51691 "EHLO tyo162.gate.nec.co.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750767AbdH1Hpa (ORCPT ); Mon, 28 Aug 2017 03:45:30 -0400 From: Naoya Horiguchi To: Baoquan He CC: Kees Cook , "linux-kernel@vger.kernel.org" , "x86@kernel.org" , "Thomas Gleixner" , "H. Peter Anvin" , Ingo Molnar , "izumi.taku@jp.fujitsu.com" , Thomas Garnier , "fanc.fnst@cn.fujitsu.com" , Matt Fleming , "Junichi Nomura" , Ard Biesheuvel Subject: [PATCH v5] x86/boot/KASLR: exclude EFI_BOOT_SERVICES_* and EFI_LOADER_* from KASLR's choice Thread-Topic: [PATCH v5] x86/boot/KASLR: exclude EFI_BOOT_SERVICES_* and EFI_LOADER_* from KASLR's choice Thread-Index: AQHTH9GEN/edhyqGMkGicEJ7FhQv7A== Date: Mon, 28 Aug 2017 07:44:45 +0000 Message-ID: <20170828074444.GC23181@hori1.linux.bs1.fc.nec.co.jp> References: <1503570831-9683-1-git-send-email-n-horiguchi@ah.jp.nec.com> <20170828065951.GB23855@x1> In-Reply-To: <20170828065951.GB23855@x1> Accept-Language: en-US, ja-JP Content-Language: ja-JP X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.128.101.13] Content-Type: text/plain; charset="iso-2022-jp" Content-ID: <59C5B3ECDF39A2469359EB0220B60CDC@gisp.nec.co.jp> Content-Transfer-Encoding: 8BIT MIME-Version: 1.0 X-TM-AS-MML: disable Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 7270 Lines: 190 On Mon, Aug 28, 2017 at 02:59:51PM +0800, Baoquan He wrote: > Hi Naoya, > > Thanks for this fix. I saw NEC had reported a bug to rhel previously, > and the bug truly will corrupt OS, it can be fixed by this patch. > > This patch looks good to me, just a small concern, please see below > inline comment. > > On 08/24/17 at 07:33pm, Naoya Horiguchi wrote: > > KASLR chooses kernel location from E820_TYPE_RAM regions by walking over > > e820 entries now. E820_TYPE_RAM includes EFI_BOOT_SERVICES_CODE and > > EFI_BOOT_SERVICES_DATA, so those regions can be the target. According to > > UEFI spec, all memory regions marked as EfiBootServicesCode and > > EfiBootServicesData are available for free memory after the first call > > of ExitBootServices(). So such regions should be usable for kernel on > > spec basis. > > > > In x86, however, we have some workaround for broken firmware, where we > > keep such regions reserved until SetVirtualAddressMap() is done. > > See the following code in should_map_region(): > > > > static bool should_map_region(efi_memory_desc_t *md) > > { > > ... > > /* > > * Map boot services regions as a workaround for buggy > > * firmware that accesses them even when they shouldn't. > > * > > * See efi_{reserve,free}_boot_services(). > > */ > > if (md->type == EFI_BOOT_SERVICES_CODE || > > md->type == EFI_BOOT_SERVICES_DATA) > > return false; > > > > This workaround suppressed a boot crash, but potential issues still > > remain because no one prevents the regions from overlapping with kernel > > image by KASLR. > > > > So let's make sure that EFI_BOOT_SERVICES_{CODE|DATA} regions are never > > chosen as kernel memory for the workaround to work fine. Furthermore, > > we choose kernel address only from EFI_CONVENTIONAL_MEMORY because it's > > the only memory type we know to be free. > > Here, I think it's better to present why EFI_CONVENTIONAL_MEMORY is the > only memory type we should choose. EFI_BOOT_SERVICES_xxx has been clear > to us why it's not good. It might be worth to saying something about > EFI_LOADER_xxx why it's not ok to choose. Maybe one sentence to mention > it and take pgd as exampel as Matt ever said. According to EFI spec, EFI_LOADER_* regions can be used for some purposes after ExitBootServices(), which is the reason for us to exclude them. Table 26. Memory Type Usage after ExitBootServices() ... EfiLoaderCode The UEFI OS Loader and/or OS may use this memory as they see fit. Note: the UEFI OS loader that called EFI_BOOT_SERVICES.ExitBootServices() is utilizing one or more EfiLoaderCode ranges. EfiLoaderData The Loader and/or OS may use this memory as they see fit. Note: the OS loader that called ExitBootServices() is utilizing one or more EfiLoaderData ranges. ... EfiBootServicesCode Memory available for general use. EfiBootServicesData Memory available for general use. EfiConventionalMemory Memory available for general use. Here is an updated patch. Thanks, Naoya Horiguchi --- >From 93182d1d8c2ce11232f7686c01dc16ee96e062c4 Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Mon, 28 Aug 2017 16:30:59 +0900 Subject: [PATCH v5] x86/boot/KASLR: exclude EFI_BOOT_SERVICES_* and EFI_LOADER_* from KASLR's choice KASLR chooses kernel location from E820_TYPE_RAM regions by walking over e820 entries now. E820_TYPE_RAM includes EFI_BOOT_SERVICES_CODE and EFI_BOOT_SERVICES_DATA, so those regions can be the target. According to UEFI spec, all memory regions marked as EfiBootServicesCode and EfiBootServicesData are available for free memory after the first call of ExitBootServices(). So such regions should be usable for kernel on spec basis. In x86, however, we have some workaround for broken firmware, where we keep such regions reserved until SetVirtualAddressMap() is done. See the following code in should_map_region(): static bool should_map_region(efi_memory_desc_t *md) { ... /* * Map boot services regions as a workaround for buggy * firmware that accesses them even when they shouldn't. * * See efi_{reserve,free}_boot_services(). */ if (md->type == EFI_BOOT_SERVICES_CODE || md->type == EFI_BOOT_SERVICES_DATA) return false; This workaround suppressed a boot crash, but potential issues still remain because no one prevents the regions from overlapping with kernel image by KASLR. So let's make sure that EFI_BOOT_SERVICES_{CODE|DATA} regions are never chosen as kernel memory for the workaround to work fine. Furthermore, EFI_LOADER_{CODE|DATA} regions are also excluded because they can be used after ExitBootServices() as defined in EFI spec. As a result, we choose kernel address only from EFI_CONVENTIONAL_MEMORY which is the only memory type we know to be free. Signed-off-by: Naoya Horiguchi --- v4 -> v5: - addressed why EFI_LOADER_* is excluded. - changed patch subject. v3 -> v4: - update comment and patch description to mention why only EFI_CONVENTIONAL_MEMORY is chosen. - use efi_early_memdesc_ptr() - I decided not to post cleanup patches (patch 2/2 in previous series) because it's not necessary to fix the issue. v2 -> v3: - skip EFI_LOADER_CODE and EFI_LOADER_DATA in region scan v1 -> v2: - switch efi_mirror_found to local variable - insert break when EFI_MEMORY_MORE_RELIABLE found --- arch/x86/boot/compressed/kaslr.c | 35 ++++++++++++++++++++++++++--------- 1 file changed, 26 insertions(+), 9 deletions(-) diff --git a/arch/x86/boot/compressed/kaslr.c b/arch/x86/boot/compressed/kaslr.c index 7de23bb279ce..ba5e9e5aaa89 100644 --- a/arch/x86/boot/compressed/kaslr.c +++ b/arch/x86/boot/compressed/kaslr.c @@ -597,19 +597,36 @@ process_efi_entries(unsigned long minimum, unsigned long image_size) for (i = 0; i < nr_desc; i++) { md = efi_early_memdesc_ptr(pmap, e->efi_memdesc_size, i); if (md->attribute & EFI_MEMORY_MORE_RELIABLE) { - region.start = md->phys_addr; - region.size = md->num_pages << EFI_PAGE_SHIFT; - process_mem_region(®ion, minimum, image_size); efi_mirror_found = true; - - if (slot_area_index == MAX_SLOT_AREA) { - debug_putstr("Aborted EFI scan (slot_areas full)!\n"); - break; - } + break; } } - return efi_mirror_found; + for (i = 0; i < nr_desc; i++) { + md = efi_early_memdesc_ptr(pmap, e->efi_memdesc_size, i); + + /* + * According to spec, EFI_BOOT_SERVICES_{CODE|DATA} are also + * available for kernel image, but we don't include them for + * the workaround for buggy firmware. + * Only EFI_CONVENTIONAL_MEMORY is guaranteed to be free. + */ + if (md->type != EFI_CONVENTIONAL_MEMORY) + continue; + + if (efi_mirror_found && + !(md->attribute & EFI_MEMORY_MORE_RELIABLE)) + continue; + + region.start = md->phys_addr; + region.size = md->num_pages << EFI_PAGE_SHIFT; + process_mem_region(®ion, minimum, image_size); + if (slot_area_index == MAX_SLOT_AREA) { + debug_putstr("Aborted EFI scan (slot_areas full)!\n"); + break; + } + } + return true; } #else static inline bool -- 2.7.4