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 70ADDECAAD4 for ; Sat, 27 Aug 2022 02:10:40 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232095AbiH0CKg (ORCPT ); Fri, 26 Aug 2022 22:10:36 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34618 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S242913AbiH0CJn (ORCPT ); Fri, 26 Aug 2022 22:09:43 -0400 Received: from dfw.source.kernel.org (dfw.source.kernel.org [IPv6:2604:1380:4641:c500::1]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id ECEA2EE495 for ; Fri, 26 Aug 2022 19:09:41 -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 800F661D14 for ; Sat, 27 Aug 2022 02:09:41 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id D2279C433D7; Sat, 27 Aug 2022 02:09:40 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linux-foundation.org; s=korg; t=1661566180; bh=5DDNriz67/FYp1x2aN9UxEN3c9aJq7pBYrO1dNbNtSY=; h=Date:To:From:Subject:From; b=Nj4J1WxzBaAxQoSuiAUcBMmj0JlmsLP4w5gl5C4n4GWfGdP/z6Wocl7WUe8ctpCqj 17Td4Fs0OyDZR0KMYvSqSdPU8NdL+9fxXG+RAz4QApsP8F//IuGVY7kdtWMHPeubZz XwHBGGNYP3AC9QdhRp9BJW8P30sAjW/7VE1E1T7E= Date: Fri, 26 Aug 2022 19:09:40 -0700 To: mm-commits@vger.kernel.org, willy@infradead.org, viro@zeniv.linux.org.uk, vegard.nossum@oracle.com, vbabka@suse.cz, tglx@linutronix.de, rostedt@goodmis.org, rientjes@google.com, pmladek@suse.com, peterz@infradead.org, penberg@kernel.org, mst@redhat.com, mingo@redhat.com, mark.rutland@arm.com, luto@kernel.org, keescook@chromium.org, iii@linux.ibm.com, iamjoonsoo.kim@lge.com, herbert@gondor.apana.org.au, hch@lst.de, gregkh@linuxfoundation.org, gor@linux.ibm.com, elver@google.com, edumazet@google.com, ebiggers@google.com, dvyukov@google.com, cl@linux.com, bp@alien8.de, axboe@kernel.dk, ast@kernel.org, arnd@arndb.de, andreyknvl@google.com, andreyknvl@gmail.com, glider@google.com, akpm@linux-foundation.org From: Andrew Morton Subject: + instrumentedh-add-kmsan-support.patch added to mm-unstable branch Message-Id: <20220827020940.D2279C433D7@smtp.kernel.org> Precedence: bulk Reply-To: linux-kernel@vger.kernel.org List-ID: X-Mailing-List: mm-commits@vger.kernel.org The patch titled Subject: instrumented.h: add KMSAN support has been added to the -mm mm-unstable branch. Its filename is instrumentedh-add-kmsan-support.patch This patch will shortly appear at https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/instrumentedh-add-kmsan-support.patch This patch will later appear in the mm-unstable branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/process/submit-checklist.rst when testing your code *** The -mm tree is included into linux-next via the mm-everything branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm and is updated there every 2-3 working days ------------------------------------------------------ From: Alexander Potapenko Subject: instrumented.h: add KMSAN support Date: Fri, 26 Aug 2022 17:07:41 +0200 To avoid false positives, KMSAN needs to unpoison the data copied from the userspace. To detect infoleaks - check the memory buffer passed to copy_to_user(). Link: https://lkml.kernel.org/r/20220826150807.723137-19-glider@google.com Signed-off-by: Alexander Potapenko Reviewed-by: Marco Elver Cc: Alexander Viro Cc: Alexei Starovoitov Cc: Andrey Konovalov Cc: Andrey Konovalov Cc: Andy Lutomirski Cc: Arnd Bergmann Cc: Borislav Petkov Cc: Christoph Hellwig Cc: Christoph Lameter Cc: David Rientjes Cc: Dmitry Vyukov Cc: Eric Biggers Cc: Eric Dumazet Cc: Greg Kroah-Hartman Cc: Herbert Xu Cc: Ilya Leoshkevich Cc: Ingo Molnar Cc: Jens Axboe Cc: Joonsoo Kim Cc: Kees Cook Cc: Mark Rutland Cc: Matthew Wilcox Cc: Michael S. Tsirkin Cc: Pekka Enberg Cc: Peter Zijlstra Cc: Petr Mladek Cc: Steven Rostedt Cc: Thomas Gleixner Cc: Vasily Gorbik Cc: Vegard Nossum Cc: Vlastimil Babka Signed-off-by: Andrew Morton --- include/linux/instrumented.h | 17 +++++++++++++- include/linux/kmsan-checks.h | 19 ++++++++++++++++ mm/kmsan/hooks.c | 38 +++++++++++++++++++++++++++++++++ 3 files changed, 73 insertions(+), 1 deletion(-) --- a/include/linux/instrumented.h~instrumentedh-add-kmsan-support +++ a/include/linux/instrumented.h @@ -2,7 +2,7 @@ /* * This header provides generic wrappers for memory access instrumentation that - * the compiler cannot emit for: KASAN, KCSAN. + * the compiler cannot emit for: KASAN, KCSAN, KMSAN. */ #ifndef _LINUX_INSTRUMENTED_H #define _LINUX_INSTRUMENTED_H @@ -10,6 +10,7 @@ #include #include #include +#include #include /** @@ -117,6 +118,7 @@ instrument_copy_to_user(void __user *to, { kasan_check_read(from, n); kcsan_check_read(from, n); + kmsan_copy_to_user(to, from, n, 0); } /** @@ -151,6 +153,19 @@ static __always_inline void instrument_copy_from_user_after(const void *to, const void __user *from, unsigned long n, unsigned long left) { + kmsan_unpoison_memory(to, n - left); } +#define instrument_get_user(to) \ +({ \ + u64 __tmp = (u64)(to); \ + kmsan_unpoison_memory(&__tmp, sizeof(__tmp)); \ + to = __tmp; \ +}) + +#define instrument_put_user(from, ptr, size) \ +({ \ + kmsan_copy_to_user(ptr, &from, sizeof(from), 0); \ +}) + #endif /* _LINUX_INSTRUMENTED_H */ --- a/include/linux/kmsan-checks.h~instrumentedh-add-kmsan-support +++ a/include/linux/kmsan-checks.h @@ -46,6 +46,21 @@ void kmsan_unpoison_memory(const void *a */ void kmsan_check_memory(const void *address, size_t size); +/** + * kmsan_copy_to_user() - Notify KMSAN about a data transfer to userspace. + * @to: destination address in the userspace. + * @from: source address in the kernel. + * @to_copy: number of bytes to copy. + * @left: number of bytes not copied. + * + * If this is a real userspace data transfer, KMSAN checks the bytes that were + * actually copied to ensure there was no information leak. If @to belongs to + * the kernel space (which is possible for compat syscalls), KMSAN just copies + * the metadata. + */ +void kmsan_copy_to_user(void __user *to, const void *from, size_t to_copy, + size_t left); + #else static inline void kmsan_poison_memory(const void *address, size_t size, @@ -58,6 +73,10 @@ static inline void kmsan_unpoison_memory static inline void kmsan_check_memory(const void *address, size_t size) { } +static inline void kmsan_copy_to_user(void __user *to, const void *from, + size_t to_copy, size_t left) +{ +} #endif --- a/mm/kmsan/hooks.c~instrumentedh-add-kmsan-support +++ a/mm/kmsan/hooks.c @@ -204,6 +204,44 @@ void kmsan_iounmap_page_range(unsigned l kmsan_leave_runtime(); } +void kmsan_copy_to_user(void __user *to, const void *from, size_t to_copy, + size_t left) +{ + unsigned long ua_flags; + + if (!kmsan_enabled || kmsan_in_runtime()) + return; + /* + * At this point we've copied the memory already. It's hard to check it + * before copying, as the size of actually copied buffer is unknown. + */ + + /* copy_to_user() may copy zero bytes. No need to check. */ + if (!to_copy) + return; + /* Or maybe copy_to_user() failed to copy anything. */ + if (to_copy <= left) + return; + + ua_flags = user_access_save(); + if ((u64)to < TASK_SIZE) { + /* This is a user memory access, check it. */ + kmsan_internal_check_memory((void *)from, to_copy - left, to, + REASON_COPY_TO_USER); + } else { + /* Otherwise this is a kernel memory access. This happens when a + * compat syscall passes an argument allocated on the kernel + * stack to a real syscall. + * Don't check anything, just copy the shadow of the copied + * bytes. + */ + kmsan_internal_memmove_metadata((void *)to, (void *)from, + to_copy - left); + } + user_access_restore(ua_flags); +} +EXPORT_SYMBOL(kmsan_copy_to_user); + /* Functions from kmsan-checks.h follow. */ void kmsan_poison_memory(const void *address, size_t size, gfp_t flags) { _ Patches currently in -mm which might be from glider@google.com are stackdepot-reserve-5-extra-bits-in-depot_stack_handle_t.patch instrumentedh-allow-instrumenting-both-sides-of-copy_from_user.patch x86-asm-instrument-usercopy-in-get_user-and-put_user.patch asm-generic-instrument-usercopy-in-cacheflushh.patch kmsan-add-rest-documentation.patch kmsan-introduce-__no_sanitize_memory-and-__no_kmsan_checks.patch kmsan-mark-noinstr-as-__no_sanitize_memory.patch x86-kmsan-pgtable-reduce-vmalloc-space.patch libnvdimm-pfn_dev-increase-max_struct_page_size.patch kmsan-add-kmsan-runtime-core.patch kmsan-disable-instrumentation-of-unsupported-common-kernel-code.patch maintainers-add-entry-for-kmsan.patch mm-kmsan-maintain-kmsan-metadata-for-page-operations.patch mm-kmsan-call-kmsan-hooks-from-slub-code.patch kmsan-handle-task-creation-and-exiting.patch init-kmsan-call-kmsan-initialization-routines.patch instrumentedh-add-kmsan-support.patch kmsan-unpoison-tlb-in-arch_tlb_gather_mmu.patch kmsan-add-iomap-support.patch input-libps2-mark-data-received-in-__ps2_command-as-initialized.patch dma-kmsan-unpoison-dma-mappings.patch virtio-kmsan-check-unpoison-scatterlist-in-vring_map_one_sg.patch kmsan-handle-memory-sent-to-from-usb.patch kmsan-add-tests-for-kmsan.patch kmsan-disable-strscpy-optimization-under-kmsan.patch crypto-kmsan-disable-accelerated-configs-under-kmsan.patch kmsan-disable-physical-page-merging-in-biovec.patch block-kmsan-skip-bio-block-merging-logic-for-kmsan.patch kcov-kmsan-unpoison-area-list-in-kcov_remote_area_put.patch security-kmsan-fix-interoperability-with-auto-initialization.patch objtool-kmsan-list-kmsan-api-functions-as-uaccess-safe.patch x86-kmsan-disable-instrumentation-of-unsupported-code.patch x86-kmsan-skip-shadow-checks-in-__switch_to.patch x86-kmsan-handle-open-coded-assembly-in-lib-iomemc.patch x86-kmsan-use-__msan_-string-functions-where-possible.patch x86-kmsan-sync-metadata-pages-on-page-fault.patch x86-kasan-kmsan-support-config_generic_csum-on-x86-enable-it-for-kasan-kmsan.patch x86-fs-kmsan-disable-config_dcache_word_access.patch x86-kmsan-dont-instrument-stack-walking-functions.patch entry-kmsan-introduce-kmsan_unpoison_entry_regs.patch bpf-kmsan-initialize-bpf-registers-with-zeroes.patch mm-fs-initialize-fsdata-passed-to-write_begin-write_end-interface.patch x86-kmsan-enable-kmsan-builds-for-x86.patch