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 A3192C433F5 for ; Tue, 30 Nov 2021 11:47:16 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S241197AbhK3Lue (ORCPT ); Tue, 30 Nov 2021 06:50:34 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45196 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236159AbhK3Ltv (ORCPT ); Tue, 30 Nov 2021 06:49:51 -0500 Received: from mail-wm1-x349.google.com (mail-wm1-x349.google.com [IPv6:2a00:1450:4864:20::349]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id CDDF0C0619D5 for ; Tue, 30 Nov 2021 03:46:05 -0800 (PST) Received: by mail-wm1-x349.google.com with SMTP id b142-20020a1c8094000000b0033f27b76819so7430857wmd.4 for ; Tue, 30 Nov 2021 03:46:05 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=date:in-reply-to:message-id:mime-version:references:subject:from:to :cc; bh=ykQMG7f7HitTY02bjgroWOqQwac8nTYOVAZLqTXm/og=; b=qfKYXOVzY2MTzD2zIwUpOpLCYIanN5AfAyzY4pD6y/00NCNsZhMgCXyQuHgqHRjMQP 2Q4GpB2kKiA9NG+mxwbdwhx/T6uztpsr0cPDZKwOOV4NBkpo8wKmqUMwj6QIZ50+WTGC rj+XujQ6x15SViviNrLc8ryKZTpDmfQ59NmqPKusALt7vIXW09vUdxRhx/IHgX0O1Fi0 peXM6+tWJqwR46RttMZspmzxuv6mgkZjTPQRNEx0+Evh/SfhzkOebPUPea948l/PAD60 8d8gkegUNMa+NHH1FPVvEqF67z8VdDW4GAXutQLRg58SDKDe/Mh9BClDvEsEpsff6SiM sbTA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:date:in-reply-to:message-id:mime-version :references:subject:from:to:cc; bh=ykQMG7f7HitTY02bjgroWOqQwac8nTYOVAZLqTXm/og=; b=DKKv0eXxjI9IY2rlk8q1zgGyOF9YzWxNeEF51FRVFlN1RpKCi717ruTD+uReZdmKHH blnG+wD3WEoKDShLF1PInFBB/wZH8mcUndYDcKhgHvVTXONNVaHFlXmpR12WdFOIzw6q OfZK3wD7Mz8DWIAmGFjCm3i+nyYROoPGVzEK1RaiLHtdK8Z8lLfrD+ATo8Zc8iH9WSDT BSg3swcjfzYWl/LQdaA2yA2HvI04z7xBJVsLPAmT3u1hrORdHj7zV/iOqW6RsZETD6JE NMyFjkrvzFi7OmP/sx7JSALpdrjrswH2BiTWNa4EZ92/kZ65tSHcUfEyFnNxKE5U9wns o5ow== X-Gm-Message-State: AOAM530sSvO+bVEXGH5WHgOQQE4xxP1vyqzy9G4lxkqPb7VYrqdWlZOd Y2VGJqSC4qeL19Exfim8LB5j7D1Z3w== X-Google-Smtp-Source: ABdhPJw6whH8Hls3EC+BgAy91BvIDfMtsebOG9cJCkHwE7EfRIVxArLwzYtCPhRhlBjkro/EFip+BpGtoQ== X-Received: from elver.muc.corp.google.com ([2a00:79e0:15:13:86b7:11e9:7797:99f0]) (user=elver job=sendgmr) by 2002:a05:600c:4f0b:: with SMTP id l11mr626212wmq.0.1638272763966; Tue, 30 Nov 2021 03:46:03 -0800 (PST) Date: Tue, 30 Nov 2021 12:44:33 +0100 In-Reply-To: <20211130114433.2580590-1-elver@google.com> Message-Id: <20211130114433.2580590-26-elver@google.com> Mime-Version: 1.0 References: <20211130114433.2580590-1-elver@google.com> X-Mailer: git-send-email 2.34.0.rc2.393.gf8c9666880-goog Subject: [PATCH v3 25/25] kcsan: Support WEAK_MEMORY with Clang where no objtool support exists From: Marco Elver To: elver@google.com, "Paul E. McKenney" Cc: Alexander Potapenko , Boqun Feng , Borislav Petkov , Dmitry Vyukov , Ingo Molnar , Mark Rutland , Peter Zijlstra , Thomas Gleixner , Waiman Long , Will Deacon , kasan-dev@googlegroups.com, linux-arch@vger.kernel.org, linux-doc@vger.kernel.org, linux-kbuild@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, llvm@lists.linux.dev, x86@kernel.org Content-Type: text/plain; charset="UTF-8" Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Clang and GCC behave a little differently when it comes to the __no_sanitize_thread attribute, which has valid reasons, and depending on context either one could be right. Traditionally, user space ThreadSanitizer [1] still expects instrumented builtin atomics (to avoid false positives) and __tsan_func_{entry,exit} (to generate meaningful stack traces), even if the function has the attribute no_sanitize("thread"). [1] https://clang.llvm.org/docs/ThreadSanitizer.html#attribute-no-sanitize-thread GCC doesn't follow the same policy (for better or worse), and removes all kinds of instrumentation if no_sanitize is added. Arguably, since this may be a problem for user space ThreadSanitizer, we expect this may change in future. Since KCSAN != ThreadSanitizer, the likelihood of false positives even without barrier instrumentation everywhere, is much lower by design. At least for Clang, however, to fully remove all sanitizer instrumentation, we must add the disable_sanitizer_instrumentation attribute, which is available since Clang 14.0. Signed-off-by: Marco Elver --- v3: * New patch. --- include/linux/compiler_types.h | 13 ++++++++++++- lib/Kconfig.kcsan | 2 +- 2 files changed, 13 insertions(+), 2 deletions(-) diff --git a/include/linux/compiler_types.h b/include/linux/compiler_types.h index 1d32f4c03c9e..3c1795fdb568 100644 --- a/include/linux/compiler_types.h +++ b/include/linux/compiler_types.h @@ -198,9 +198,20 @@ struct ftrace_likely_data { # define __no_kasan_or_inline __always_inline #endif -#define __no_kcsan __no_sanitize_thread #ifdef __SANITIZE_THREAD__ +/* + * Clang still emits instrumentation for __tsan_func_{entry,exit}() and builtin + * atomics even with __no_sanitize_thread (to avoid false positives in userspace + * ThreadSanitizer). The kernel's requirements are stricter and we really do not + * want any instrumentation with __no_kcsan. + * + * Therefore we add __disable_sanitizer_instrumentation where available to + * disable all instrumentation. See Kconfig.kcsan where this is mandatory. + */ +# define __no_kcsan __no_sanitize_thread __disable_sanitizer_instrumentation # define __no_sanitize_or_inline __no_kcsan notrace __maybe_unused +#else +# define __no_kcsan #endif #ifndef __no_sanitize_or_inline diff --git a/lib/Kconfig.kcsan b/lib/Kconfig.kcsan index e4394ea8068b..63b70b8c5551 100644 --- a/lib/Kconfig.kcsan +++ b/lib/Kconfig.kcsan @@ -198,7 +198,7 @@ config KCSAN_WEAK_MEMORY # We can either let objtool nop __tsan_func_{entry,exit}() and builtin # atomics instrumentation in .noinstr.text, or use a compiler that can # implement __no_kcsan to really remove all instrumentation. - depends on STACK_VALIDATION || CC_IS_GCC + depends on STACK_VALIDATION || CC_IS_GCC || CLANG_VERSION >= 140000 help Enable support for modeling a subset of weak memory, which allows detecting a subset of data races due to missing memory barriers. -- 2.34.0.rc2.393.gf8c9666880-goog