From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-qk1-f180.google.com (mail-qk1-f180.google.com [209.85.222.180]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 9CC921FB4 for ; Thu, 3 Oct 2024 00:08:45 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.222.180 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1727914128; cv=none; b=uLFwjVzXBPlgsHRlIf3QLCwX9nYgr5bRMtZ8mtlgAmuR5ttVNRqwGZcbyBBPd7ScxxfQEaxKhVltm9HIWpuApkGSVvHdCqg/XkJlIk3OIOLJsy2SCj9jtwVTuX7F5PFUI2J7zFgKCv4akM/CG2C8DDLn7IB0FeIWHqcbqpOFWg4= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1727914128; c=relaxed/simple; bh=KUrfk+eq3257BUq5EaY4utlmEACyrb2EHZYgj/5cUlI=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version: Content-Type:Content-Disposition:In-Reply-To; b=EKi8pJ+NGeMs66/uqKjvcMb3au7mFK2r00nMSiGprMN9Vu6U7PXmJgTp+elTwbwcHYJVj29am8nUS4/AUBTlt315Om0bqEb0u2oZtGKriwe3j0Z0oxP27fxkQ3JoVHQhqnWq3NG5qGySP/Uic4auMof7r+lBOsgP+wjPYy5FzRc= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=joelfernandes.org; spf=pass smtp.mailfrom=joelfernandes.org; dkim=pass (1024-bit key) header.d=joelfernandes.org header.i=@joelfernandes.org header.b=FcPJMPyh; arc=none smtp.client-ip=209.85.222.180 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=joelfernandes.org Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=joelfernandes.org Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=joelfernandes.org header.i=@joelfernandes.org header.b="FcPJMPyh" Received: by mail-qk1-f180.google.com with SMTP id af79cd13be357-7a9ae0e116cso38073985a.1 for ; Wed, 02 Oct 2024 17:08:45 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelfernandes.org; s=google; t=1727914124; x=1728518924; darn=lists.linux.dev; h=in-reply-to:content-transfer-encoding:content-disposition :mime-version:references:message-id:subject:cc:to:from:date:from:to :cc:subject:date:message-id:reply-to; bh=d3cwmVH4T8IqaLZrJPf4dN6HB0iRJ3ayFzGdEtNHxoA=; b=FcPJMPyhq5Y7DkuOCSJOGdEms8WTSmAr3+EYeLQ5tIgWnwtgOCU0UL7y2z227TaitF HsznbtY5WkMz3S8C4tRGsRL2dG9I0Xijsgf9VjE8tgOegLK3r34IOHZ06ljjIYlmsHaw fu0E8JoarQBIIG2oyD8YM0Ruk5NyDeCYiDotI= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1727914124; x=1728518924; h=in-reply-to:content-transfer-encoding:content-disposition :mime-version:references:message-id:subject:cc:to:from:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=d3cwmVH4T8IqaLZrJPf4dN6HB0iRJ3ayFzGdEtNHxoA=; b=mo6CJqaKbZXC0Ofvd8OkIZdM6D2VQowImp8h+Oo+ZOum0pTXxXlbirfQt1/xSTLbH3 moeSW/vWSa3r5rfQuoP9PrQRiugr308v9wSYEcsJHGB5F0DrhxCj6AMPWSbB2RmrVbFQ 7YsaAzYAdcfLZZafBKFe0dML3izT+JGgsFHJDvx564eqmpsDgCebBT1U1VMXc6Ds6UGO njo19BmKNBG6OpHX2R3cvEKbMPS+BkZrSjVmnnUJcrP4NBmDuC4wuWfrDV8ecydaaVG/ 3wcDBFkVvoZtwEjJQL5ITb2LJYVPEAu66K2Pk9MAEbtSNqGTblbKmytyQnbZqsRWO7iC a38g== X-Forwarded-Encrypted: i=1; AJvYcCVu0zqhUka9+ujhQMyowQGDhzdE83Sp2NEBCNETQ5r8QwbT88OscFL5AIKSa+VR9RwlTmoY@lists.linux.dev X-Gm-Message-State: AOJu0Yw8LZefKrARaMial2ZhsYtWs7xfKv6algoeOdoK+055UK/BfuPU g1KM+qj/iF5yBFe+cOyTZa6/LfKqUL+gP1Q/VE8QDwBOXcs/cVGvCu0o/f6+J5Y= X-Google-Smtp-Source: AGHT+IFyyGM2DUoFUWQ5HNQaWWB8/IaPvClgER78DCcGgoTY7+i7cDT8H5iX51KO9rlUJfwMgd+NTQ== X-Received: by 2002:a05:620a:2697:b0:7ac:dd3a:6787 with SMTP id af79cd13be357-7ae62732decmr738226785a.52.1727914124476; Wed, 02 Oct 2024 17:08:44 -0700 (PDT) Received: from localhost (129.177.85.34.bc.googleusercontent.com. [34.85.177.129]) by smtp.gmail.com with ESMTPSA id af79cd13be357-7ae3783f80bsm669783585a.123.2024.10.02.17.08.43 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 02 Oct 2024 17:08:43 -0700 (PDT) Date: Thu, 3 Oct 2024 00:08:43 +0000 From: Joel Fernandes To: Mathieu Desnoyers Cc: Linus Torvalds , Andrew Morton , Peter Zijlstra , linux-kernel@vger.kernel.org, Nicholas Piggin , Michael Ellerman , Greg Kroah-Hartman , Sebastian Andrzej Siewior , "Paul E. McKenney" , Will Deacon , Boqun Feng , Alan Stern , John Stultz , Neeraj Upadhyay , Frederic Weisbecker , Josh Triplett , Uladzislau Rezki , Steven Rostedt , Lai Jiangshan , Zqiang , Ingo Molnar , Waiman Long , Mark Rutland , Thomas Gleixner , Vlastimil Babka , maged.michael@gmail.com, Mateusz Guzik , Jonas Oberhauser , rcu@vger.kernel.org, linux-mm@kvack.org, lkmm@lists.linux.dev, Gary Guo , Nikita Popov , llvm@lists.linux.dev Subject: Re: [RFC PATCH 1/4] compiler.h: Introduce ptr_eq() to preserve address dependency Message-ID: <20241003000843.GA192403@google.com> References: <20241002010205.1341915-1-mathieu.desnoyers@efficios.com> <20241002010205.1341915-2-mathieu.desnoyers@efficios.com> Precedence: bulk X-Mailing-List: llvm@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20241002010205.1341915-2-mathieu.desnoyers@efficios.com> On Tue, Oct 01, 2024 at 09:02:02PM -0400, Mathieu Desnoyers wrote: > Compiler CSE and SSA GVN optimizations can cause the address dependency > of addresses returned by rcu_dereference to be lost when comparing those > pointers with either constants or previously loaded pointers. > > Introduce ptr_eq() to compare two addresses while preserving the address > dependencies for later use of the address. It should be used when > comparing an address returned by rcu_dereference(). > > This is needed to prevent the compiler CSE and SSA GVN optimizations > from using @a (or @b) in places where the source refers to @b (or @a) > based on the fact that after the comparison, the two are known to be > equal, which does not preserve address dependencies and allows the > following misordering speculations: > > - If @b is a constant, the compiler can issue the loads which depend > on @a before loading @a. > - If @b is a register populated by a prior load, weakly-ordered > CPUs can speculate loads which depend on @a before loading @a. > > The same logic applies with @a and @b swapped. > [...] > +/* > + * Compare two addresses while preserving the address dependencies for > + * later use of the address. It should be used when comparing an address > + * returned by rcu_dereference(). > + * > + * This is needed to prevent the compiler CSE and SSA GVN optimizations > + * from using @a (or @b) in places where the source refers to @b (or @a) > + * based on the fact that after the comparison, the two are known to be > + * equal, which does not preserve address dependencies and allows the > + * following misordering speculations: > + * > + * - If @b is a constant, the compiler can issue the loads which depend > + * on @a before loading @a. > + * - If @b is a register populated by a prior load, weakly-ordered > + * CPUs can speculate loads which depend on @a before loading @a. > + * > + * The same logic applies with @a and @b swapped. > + * > + * Return value: true if pointers are equal, false otherwise. > + * > + * The compiler barrier() is ineffective at fixing this issue. It does > + * not prevent the compiler CSE from losing the address dependency: > + * > + * int fct_2_volatile_barriers(void) > + * { > + * int *a, *b; > + * > + * do { > + * a = READ_ONCE(p); > + * asm volatile ("" : : : "memory"); > + * b = READ_ONCE(p); > + * } while (a != b); > + * asm volatile ("" : : : "memory"); <-- barrier() > + * return *b; > + * } > + * > + * With gcc 14.2 (arm64): > + * > + * fct_2_volatile_barriers: > + * adrp x0, .LANCHOR0 > + * add x0, x0, :lo12:.LANCHOR0 > + * .L2: > + * ldr x1, [x0] <-- x1 populated by first load. > + * ldr x2, [x0] > + * cmp x1, x2 > + * bne .L2 > + * ldr w0, [x1] <-- x1 is used for access which should depend on b. > + * ret > + * I could reproduce this in compiler explorer, but I'm curious what flags are you using? For me it does a bunch of usage of the stack for temporary storage (still incorrectly returns *a though as you pointed). Interestingly, if I just move the comparison into an an __always_inline__ function like below, but without the optimizer hide stuff, gcc 14.2 on arm64 does generate the correct code: static inline __attribute__((__always_inline__)) int ptr_eq(const volatile void *a, const volatile void *b) {     /* No OPTIMIZER_HIDE_VAR */     return a == b; } volatile int *p = 0; int fct_2_volatile_barriers() {     int *a, *b;     do {         a = READ_ONCE(p);         asm volatile ("" : : : "memory");         b = READ_ONCE(p);     } while (!ptr_eq(a, b));     asm volatile ("" : : : "memory");  // barrier()     return *b; } But not sure if it fixes the speculation issue you referred to. Putting back the OPTIMIZER_HIDE_VAR() then just seems to pass the a and b stored on the stack through a washing machine:         ldr     x0, [sp, 8]         str     x0, [sp, 8]         ldr     x0, [sp]         str     x0, [sp] And here I thought the "" in OPTIMIZER_HIDE_VAR was not supposed to generate any code but I guess it is still a NOOP. Anyway, as such this LGTM since whether OPTIMIZER_HIDE_VAR() used or not, it does fix the problem. Reviewed-by: Joel Fernandes (Google) thanks, - Joel