From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marco Elver Subject: [PATCH 5/8] seqlock: Require WRITE_ONCE surrounding raw_seqcount_barrier Date: Wed, 16 Oct 2019 10:39:56 +0200 Message-ID: <20191016083959.186860-6-elver@google.com> References: <20191016083959.186860-1-elver@google.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Return-path: In-Reply-To: <20191016083959.186860-1-elver@google.com> Sender: linux-kernel-owner@vger.kernel.org To: elver@google.com Cc: akiyks@gmail.com, stern@rowland.harvard.edu, glider@google.com, parri.andrea@gmail.com, andreyknvl@google.com, luto@kernel.org, ard.biesheuvel@linaro.org, arnd@arndb.de, boqun.feng@gmail.com, bp@alien8.de, dja@axtens.net, dlustig@nvidia.com, dave.hansen@linux.intel.com, dhowells@redhat.com, dvyukov@google.com, hpa@zytor.com, mingo@redhat.com, j.alglave@ucl.ac.uk, joel@joelfernandes.org, corbet@lwn.net, jpoimboe@redhat.com, luc.maranget@inria.fr, mark.rutland@arm.com, npiggin@gmail.com, paulmck@linux.ibm.com, peterz@infradead.org, tglx@linutronix.de, will@kernel.org, kasan-dev@googlegroups.com, linux-arch@vger.kernel.org, linux-doc@vger.kernel.org, linux-efi@vger.kernel.org, linux-kbuild@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, x86@kernel.org List-Id: linux-arch.vger.kernel.org This patch proposes to require marked atomic accesses surrounding raw_write_seqcount_barrier. We reason that otherwise there is no way to guarantee propagation nor atomicity of writes before/after the barrier [1]. For example, consider the compiler tears stores either before or after the barrier; in this case, readers may observe a partial value, and because readers are unaware that writes are going on (writes are not in a seq-writer critical section), will complete the seq-reader critical section while having observed some partial state. [1] https://lwn.net/Articles/793253/ This came up when designing and implementing KCSAN, because KCSAN would flag these accesses as data-races. After careful analysis, our reasoning as above led us to conclude that the best thing to do is to propose an amendment to the raw_seqcount_barrier usage. Signed-off-by: Marco Elver --- include/linux/seqlock.h | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/include/linux/seqlock.h b/include/linux/seqlock.h index 1e425831a7ed..5d50aad53b47 100644 --- a/include/linux/seqlock.h +++ b/include/linux/seqlock.h @@ -265,6 +265,13 @@ static inline void raw_write_seqcount_end(seqcount_t *s) * usual consistency guarantee. It is one wmb cheaper, because we can * collapse the two back-to-back wmb()s. * + * Note that, writes surrounding the barrier should be declared atomic (e.g. + * via WRITE_ONCE): a) to ensure the writes become visible to other threads + * atomically, avoiding compiler optimizations; b) to document which writes are + * meant to propagate to the reader critical section. This is necessary because + * neither writes before and after the barrier are enclosed in a seq-writer + * critical section that would ensure readers are aware of ongoing writes. + * * seqcount_t seq; * bool X = true, Y = false; * -- 2.23.0.700.g56cf767bdb-goog From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk1-f202.google.com ([209.85.222.202]:56909 "EHLO mail-qk1-f202.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2391786AbfJPIl1 (ORCPT ); Wed, 16 Oct 2019 04:41:27 -0400 Received: by mail-qk1-f202.google.com with SMTP id b29so2004659qka.23 for ; Wed, 16 Oct 2019 01:41:25 -0700 (PDT) Date: Wed, 16 Oct 2019 10:39:56 +0200 In-Reply-To: <20191016083959.186860-1-elver@google.com> Message-ID: <20191016083959.186860-6-elver@google.com> Mime-Version: 1.0 References: <20191016083959.186860-1-elver@google.com> Subject: [PATCH 5/8] seqlock: Require WRITE_ONCE surrounding raw_seqcount_barrier From: Marco Elver Content-Type: text/plain; charset="UTF-8" Sender: linux-arch-owner@vger.kernel.org List-ID: To: elver@google.com Cc: akiyks@gmail.com, stern@rowland.harvard.edu, glider@google.com, parri.andrea@gmail.com, andreyknvl@google.com, luto@kernel.org, ard.biesheuvel@linaro.org, arnd@arndb.de, boqun.feng@gmail.com, bp@alien8.de, dja@axtens.net, dlustig@nvidia.com, dave.hansen@linux.intel.com, dhowells@redhat.com, dvyukov@google.com, hpa@zytor.com, mingo@redhat.com, j.alglave@ucl.ac.uk, joel@joelfernandes.org, corbet@lwn.net, jpoimboe@redhat.com, luc.maranget@inria.fr, mark.rutland@arm.com, npiggin@gmail.com, paulmck@linux.ibm.com, peterz@infradead.org, tglx@linutronix.de, will@kernel.org, kasan-dev@googlegroups.com, linux-arch@vger.kernel.org, linux-doc@vger.kernel.org, linux-efi@vger.kernel.org, linux-kbuild@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, x86@kernel.org Message-ID: <20191016083956.Re264eZXyhxhCiqPbt2pJ6sb02ENU__-41oVCnzhzXs@z> This patch proposes to require marked atomic accesses surrounding raw_write_seqcount_barrier. We reason that otherwise there is no way to guarantee propagation nor atomicity of writes before/after the barrier [1]. For example, consider the compiler tears stores either before or after the barrier; in this case, readers may observe a partial value, and because readers are unaware that writes are going on (writes are not in a seq-writer critical section), will complete the seq-reader critical section while having observed some partial state. [1] https://lwn.net/Articles/793253/ This came up when designing and implementing KCSAN, because KCSAN would flag these accesses as data-races. After careful analysis, our reasoning as above led us to conclude that the best thing to do is to propose an amendment to the raw_seqcount_barrier usage. Signed-off-by: Marco Elver --- include/linux/seqlock.h | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/include/linux/seqlock.h b/include/linux/seqlock.h index 1e425831a7ed..5d50aad53b47 100644 --- a/include/linux/seqlock.h +++ b/include/linux/seqlock.h @@ -265,6 +265,13 @@ static inline void raw_write_seqcount_end(seqcount_t *s) * usual consistency guarantee. It is one wmb cheaper, because we can * collapse the two back-to-back wmb()s. * + * Note that, writes surrounding the barrier should be declared atomic (e.g. + * via WRITE_ONCE): a) to ensure the writes become visible to other threads + * atomically, avoiding compiler optimizations; b) to document which writes are + * meant to propagate to the reader critical section. This is necessary because + * neither writes before and after the barrier are enclosed in a seq-writer + * critical section that would ensure readers are aware of ongoing writes. + * * seqcount_t seq; * bool X = true, Y = false; * -- 2.23.0.700.g56cf767bdb-goog