From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from simark.ca by simark.ca with LMTP id Vg5bCQVvlGTd7woAWB0awg (envelope-from ) for ; Thu, 22 Jun 2023 11:55:49 -0400 Authentication-Results: simark.ca; dkim=pass (2048-bit key; unprotected) header.d=lists.lttng.org header.i=@lists.lttng.org header.a=rsa-sha256 header.s=default header.b=j6FH4dj5; dkim-atps=neutral Received: by simark.ca (Postfix, from userid 112) id 17B721E0BB; Thu, 22 Jun 2023 11:55:49 -0400 (EDT) Received: from lists.lttng.org (lists.lttng.org [167.114.26.123]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by simark.ca (Postfix) with ESMTPS id 8CC9D1E0AC for ; Thu, 22 Jun 2023 11:55:46 -0400 (EDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=lists.lttng.org; s=default; t=1687449346; bh=HjlCiN7l15pa1mFRRiLOmh1mlMOqqSrMho+QnIDktCw=; h=Date:To:Cc:References:In-Reply-To:Subject:List-Id: List-Unsubscribe:List-Archive:List-Post:List-Help:List-Subscribe: From:Reply-To:From; b=j6FH4dj5VyAEbl4m9hmrPKwkqtaihCuZIwcuL/+hCS16gdlnkeYRX3q3Q9fRbhaRW bqWg7yKDI5bA5h3hsbxfz9OqNfgAJ93jmln6oShAygfcZkduGuMQnj7NbSSpIuPuuq VesYuG7cvK2fZFCB51JZJqVDsF2RoIjYjdWH2ItyP40hRDoa9nVjDkmsAeThQY0Sus 0fXq6Ra+LbrRaX8mrA+cbD6+1BnMCkh0TY2OdQRXW06VfoHM0RfVZM8GKOe0fA4Dl7 nfJuFexOlN7tR9D8LivZEXKz0s9oHqEPhl51dn4Q2nFhnkLn+zyC8yyaJGCes3WS1y lDyS0OweZzBfw== Received: from lists-lttng01.efficios.com (localhost [IPv6:::1]) by lists.lttng.org (Postfix) with ESMTP id 4Qn4hx6xrlz20VC; Thu, 22 Jun 2023 11:55:45 -0400 (EDT) Received: from smtpout.efficios.com (smtpout.efficios.com [167.114.26.122]) by lists.lttng.org (Postfix) with ESMTPS id 4Qn4hw6MYXz20VB for ; Thu, 22 Jun 2023 11:55:44 -0400 (EDT) Received: from [172.16.0.134] (192-222-143-198.qc.cable.ebox.net [192.222.143.198]) by smtpout.efficios.com (Postfix) with ESMTPSA id 4Qn4hh0tSfz191S; Thu, 22 Jun 2023 11:55:32 -0400 (EDT) Message-ID: Date: Thu, 22 Jun 2023 11:55:55 -0400 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.12.0 Content-Language: en-US To: paulmck@kernel.org, Olivier Dion Cc: lttng-dev@lists.lttng.org References: <20230515201718.9809-1-odion@efficios.com> <20230515201718.9809-3-odion@efficios.com> In-Reply-To: Subject: Re: [lttng-dev] [PATCH 02/11] urcu/uatomic: Use atomic builtins if configured X-BeenThere: lttng-dev@lists.lttng.org X-Mailman-Version: 2.1.39 Precedence: list List-Id: LTTng development list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , From: Mathieu Desnoyers via lttng-dev Reply-To: Mathieu Desnoyers Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Errors-To: lttng-dev-bounces@lists.lttng.org Sender: "lttng-dev" On 6/21/23 19:19, Paul E. McKenney wrote: [...] >> diff --git a/include/urcu/uatomic/builtins-generic.h b/include/urcu/uatomic/builtins-generic.h >> new file mode 100644 >> index 0000000..8e6a9b5 >> --- /dev/null >> +++ b/include/urcu/uatomic/builtins-generic.h >> @@ -0,0 +1,85 @@ >> +/* >> + * urcu/uatomic/builtins-generic.h >> + * >> + * Copyright (c) 2023 Olivier Dion >> + * >> + * This library is free software; you can redistribute it and/or >> + * modify it under the terms of the GNU Lesser General Public >> + * License as published by the Free Software Foundation; either >> + * version 2.1 of the License, or (at your option) any later version. >> + * >> + * This library is distributed in the hope that it will be useful, >> + * but WITHOUT ANY WARRANTY; without even the implied warranty of >> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU >> + * Lesser General Public License for more details. >> + * >> + * You should have received a copy of the GNU Lesser General Public >> + * License along with this library; if not, write to the Free Software >> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA >> + */ >> + >> +#ifndef _URCU_UATOMIC_BUILTINS_GENERIC_H >> +#define _URCU_UATOMIC_BUILTINS_GENERIC_H >> + >> +#include >> + >> +#define uatomic_set(addr, v) __atomic_store_n(addr, v, __ATOMIC_RELAXED) >> + >> +#define uatomic_read(addr) __atomic_load_n(addr, __ATOMIC_RELAXED) > > Does this lose the volatile semantics that the old-style definitions > had? > Yes. [...] >> +++ b/include/urcu/uatomic/builtins-x86.h >> @@ -0,0 +1,85 @@ >> +/* >> + * urcu/uatomic/builtins-x86.h >> + * >> + * Copyright (c) 2023 Olivier Dion >> + * >> + * This library is free software; you can redistribute it and/or >> + * modify it under the terms of the GNU Lesser General Public >> + * License as published by the Free Software Foundation; either >> + * version 2.1 of the License, or (at your option) any later version. >> + * >> + * This library is distributed in the hope that it will be useful, >> + * but WITHOUT ANY WARRANTY; without even the implied warranty of >> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU >> + * Lesser General Public License for more details. >> + * >> + * You should have received a copy of the GNU Lesser General Public >> + * License along with this library; if not, write to the Free Software >> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA >> + */ >> + >> +#ifndef _URCU_UATOMIC_BUILTINS_X86_H >> +#define _URCU_UATOMIC_BUILTINS_X86_H >> + >> +#include >> + >> +#define uatomic_set(addr, v) __atomic_store_n(addr, v, __ATOMIC_RELAXED) >> + >> +#define uatomic_read(addr) __atomic_load_n(addr, __ATOMIC_RELAXED) > > And same question here. Yes, this opens interesting questions: * what semantic do we want for uatomic_read/set ? * what semantic do we want for CMM_LOAD_SHARED/CMM_STORE_SHARED ? * do we want to allow load/store-shared to work on variables larger than a word ? (e.g. on a uint64_t on a 32-bit architecture, or on a structure) * what are the guarantees of a volatile type ? * what are the guarantees of a load/store relaxed in C11 ? Does the delta between volatile and C11 relaxed guarantees matter ? Is there an advantage to use C11 load/store relaxed over volatile ? Should we combine both C11 load/store _and_ volatile ? Should we use atomic_signal_fence instead ? Thanks, Mathieu > > Thanx, Paul > >> + -- Mathieu Desnoyers EfficiOS Inc. https://www.efficios.com _______________________________________________ lttng-dev mailing list lttng-dev@lists.lttng.org https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev