From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 9774A43FDC; Wed, 8 May 2024 11:27:33 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 84A2140A6E; Wed, 8 May 2024 11:27:33 +0200 (CEST) Received: from mail.lysator.liu.se (mail.lysator.liu.se [130.236.254.3]) by mails.dpdk.org (Postfix) with ESMTP id 20585402AF for ; Wed, 8 May 2024 11:27:32 +0200 (CEST) Received: from mail.lysator.liu.se (localhost [127.0.0.1]) by mail.lysator.liu.se (Postfix) with ESMTP id 6EEB39A91 for ; Wed, 8 May 2024 11:27:31 +0200 (CEST) Received: by mail.lysator.liu.se (Postfix, from userid 1004) id 62CF69A90; Wed, 8 May 2024 11:27:31 +0200 (CEST) X-Spam-Checker-Version: SpamAssassin 4.0.0 (2022-12-13) on hermod.lysator.liu.se X-Spam-Level: X-Spam-Status: No, score=-1.3 required=5.0 tests=ALL_TRUSTED,AWL, T_SCC_BODY_TEXT_LINE autolearn=disabled version=4.0.0 X-Spam-Score: -1.3 Received: from [192.168.1.59] (h-62-63-215-114.A163.priv.bahnhof.se [62.63.215.114]) (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 mail.lysator.liu.se (Postfix) with ESMTPSA id 088189B04; Wed, 8 May 2024 11:27:28 +0200 (CEST) Message-ID: <1c557696-492d-445c-af5c-9341a538d782@lysator.liu.se> Date: Wed, 8 May 2024 11:27:28 +0200 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [RFC v7 3/6] eal: add exactly-once bit access functions To: =?UTF-8?Q?Morten_Br=C3=B8rup?= , =?UTF-8?Q?Mattias_R=C3=B6nnblom?= , dev@dpdk.org Cc: Heng Wang , Stephen Hemminger , Tyler Retzlaff References: <20240502055706.112443-2-mattias.ronnblom@ericsson.com> <20240505083737.118649-1-mattias.ronnblom@ericsson.com> <20240505083737.118649-4-mattias.ronnblom@ericsson.com> <98CBD80474FA8B44BF855DF32C47DC35E9F426@smartserver.smartshare.dk> <355b3d73-94a8-4734-8934-22081d6e7b11@lysator.liu.se> <98CBD80474FA8B44BF855DF32C47DC35E9F428@smartserver.smartshare.dk> <98CBD80474FA8B44BF855DF32C47DC35E9F429@smartserver.smartshare.dk> Content-Language: en-US From: =?UTF-8?Q?Mattias_R=C3=B6nnblom?= In-Reply-To: <98CBD80474FA8B44BF855DF32C47DC35E9F429@smartserver.smartshare.dk> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-Virus-Scanned: ClamAV using ClamSMTP X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org On 2024-05-08 10:11, Morten Brørup wrote: >> From: Mattias Rönnblom [mailto:hofors@lysator.liu.se] >> Sent: Wednesday, 8 May 2024 10.00 >> >> On 2024-05-08 09:33, Morten Brørup wrote: >>>> From: Mattias Rönnblom [mailto:hofors@lysator.liu.se] >>>> Sent: Wednesday, 8 May 2024 08.47 >>>> >>>> On 2024-05-07 21:17, Morten Brørup wrote: >>>>>> From: Mattias Rönnblom [mailto:mattias.ronnblom@ericsson.com] >>>>>> Sent: Sunday, 5 May 2024 10.38 >>>>>> >>>>>> Add test/set/clear/assign/flip functions which prevents certain >>>>>> compiler optimizations and guarantees that program-level memory loads >>>>>> and/or stores will actually occur. >>>>>> >>>>>> These functions are useful when interacting with memory-mapped >>>>>> hardware devices. >>>>>> >>>>>> The "once" family of functions does not promise atomicity and provides >>>>>> no memory ordering guarantees beyond the C11 relaxed memory model. >>>>> >>>>> In another thread, Stephen referred to the extended discussion on memory >>>> models in Linux kernel documentation: >>>>> https://www.kernel.org/doc/html/latest/core-api/wrappers/memory- >>>> barriers.html >>>>> >>>>> Unlike the "once" family of functions in this RFC, the "once" family of >>>> functions in the kernel also guarantee memory ordering, specifically for >>>> memory-mapped hardware devices. The document describes the rationale with >>>> examples. >>>>> >>>> >>>> What more specifically did you have in mind? READ_ONCE() and >>>> WRITE_ONCE()? They give almost no guarantees. Very much relaxed. >>> >>> The way I read it, they do provide memory ordering guarantees. >>> >> >> Sure. All types memory operations comes with some kind guarantees. A >> series of non-atomic, non-volatile stores issued by a particular thread >> are guaranteed to happen in program order, from the point of view of >> that thread, for example. Would be hard to write a program if that >> wasn't true. >> >> "This macro does not give any guarantees in regards to memory ordering /../" >> >> This is not true. I will rephrase to "any *additional* guarantees" for >> both plain and "once" family documentation. > > Consider code like this: > set_once(HW_START_BIT); > while (!get_once(HW_DONE_BIT)) /*busy wait*/; > > If the "once" functions are used for hardware access, they must guarantee that HW_START_BIT has been written before HW_DONE_BIT is read. > Provided bits reside in the same word, there is (or at least, should be) such guarantee, and otherwise, you'll need a barrier. I'm guessing in most cases the requirements are actually not as strict as you pose them: DONE starts as 0, so it may actually be read before START is written to, but not all DONE reads can be reordered ahead of the single START write. In that case, a compiler barrier between set and the get loop should suffice. Otherwise, you need a full barrier, or an I/O barrier. Anyway, since the exact purpose of the "once" type bit operations is unclear, maybe I should drop them from the patch set. Now, they are much like the Linux kernel's __set_bit(), but for hardware access, maybe they should be more like writel(). > The documentation must reflect this ordering guarantee. > >> >>> Ignore that the kernel's "once" functions operates on words and this RFC >> operates on bits, the behavior is the same. Either there are memory ordering >> guarantees, or there are not. >>> >>>> >>>> I've read that document. >>>> >>>> What you should keep in mind if you read that document, is that DPDK >>>> doesn't use the kernel's memory model, and doesn't have the kernel's >>>> barrier and atomics APIs. What we have are an obsolete, miniature >>>> look-alike in and something C11-like in . >>>> >>>> My general impression is that DPDK was moving in the C11 direction >>>> memory model-wise, which is not the model the kernel uses. >>> >>> I think you and I agree that using legacy methods only because "the kernel >> does it that way" would not be the optimal roadmap for DPDK. >>> >>> We should keep moving in the C11 direction memory model-wise. >>> I consider it more descriptive, and thus expect compilers to eventually >> produce better optimized code. >>> >>>> >>>>> It makes me think that DPDK "once" family of functions should behave >>>> similarly. >>>> >>>> I think they do already. >>> >>> I haven't looked deep into it, but the RFC's documentation says otherwise: >>> The "once" family of functions does not promise atomicity and provides *no >> memory ordering* guarantees beyond the C11 relaxed memory model. >>> >>>> >>>> Also, rte_bit_once_set() works as the kernel's __set_bit(). >>>> >>>>> Alternatively, if the "once" family of functions cannot be generically >>>> implemented with a memory ordering that is optimal for all use cases, drop >>>> this family of functions, and instead rely on the "atomic" family of >> functions >>>> for interacting with memory-mapped hardware devices. >>>>> >>>>>> >>>>>> RFC v7: >>>>>> * Fix various minor issues in documentation. >>>>>> >>>>>> RFC v6: >>>>>> * Have rte_bit_once_test() accept const-marked bitsets. >>>>>> >>>>>> RFC v3: >>>>>> * Work around lack of C++ support for _Generic (Tyler Retzlaff). >>>>>> >>>>>> Signed-off-by: Mattias Rönnblom >>>>>> Acked-by: Morten Brørup >>>>>> Acked-by: Tyler Retzlaff >>>>>> --- >>>>>