From mboxrd@z Thu Jan 1 00:00:00 1970 From: peterz@infradead.org Subject: Re: [PATCH 3/8] kcsan: Skew delay to be longer for certain access types Date: Tue, 21 Jul 2020 16:34:32 +0200 Message-ID: <20200721143432.GM119549@hirez.programming.kicks-ass.net> References: <20200721103016.3287832-1-elver@google.com> <20200721103016.3287832-4-elver@google.com> <20200721140523.GA10769@hirez.programming.kicks-ass.net> <20200721142654.GA3396394@elver.google.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from lindbergh.monkeyblade.net ([23.128.96.19]:55142 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726412AbgGUOel (ORCPT ); Tue, 21 Jul 2020 10:34:41 -0400 Content-Disposition: inline In-Reply-To: <20200721142654.GA3396394@elver.google.com> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Marco Elver Cc: paulmck@kernel.org, will@kernel.org, arnd@arndb.de, mark.rutland@arm.com, dvyukov@google.com, glider@google.com, kasan-dev@googlegroups.com, linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org On Tue, Jul 21, 2020 at 04:26:54PM +0200, Marco Elver wrote: > I'll rewrite the commit message: > > For compound instrumentation and assert accesses, skew the > watchpoint delay to be longer if randomized. This is useful to > improve race detection for such accesses. > > For compound accesses we should increase the delay as we've > aggregated both read and write instrumentation. By giving up 1 > call into the runtime, we're less likely to set up a watchpoint > and thus less likely to detect a race. We can balance this by > increasing the watchpoint delay. Aah, makes sense now. Thanks!