From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-2.2 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id E0642C47255 for ; Mon, 11 May 2020 11:45:21 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id BD69F20735 for ; Mon, 11 May 2020 11:45:21 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728613AbgEKLpV (ORCPT ); Mon, 11 May 2020 07:45:21 -0400 Received: from foss.arm.com ([217.140.110.172]:58526 "EHLO foss.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728531AbgEKLpV (ORCPT ); Mon, 11 May 2020 07:45:21 -0400 Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 15976101E; Mon, 11 May 2020 04:45:18 -0700 (PDT) Received: from [192.168.2.22] (unknown [172.31.20.19]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 7A59D3F305; Mon, 11 May 2020 04:45:16 -0700 (PDT) Subject: Re: [PATCH RFC 1/8] dmaengine: Actions: get rid of bit fields from dma descriptor To: Manivannan Sadhasivam , Amit Tomer Cc: vkoul@kernel.org, =?UTF-8?Q?Andreas_F=c3=a4rber?= , dan.j.williams@intel.com, cristian.ciocaltea@gmail.com, dmaengine@vger.kernel.org, linux-arm-kernel , linux-actions@lists.infradead.org References: <1588761371-9078-1-git-send-email-amittomer25@gmail.com> <1588761371-9078-2-git-send-email-amittomer25@gmail.com> <20200510155159.GA27924@Mani-XPS-13-9360> <20200511112014.GA3322@Mani-XPS-13-9360> From: =?UTF-8?Q?Andr=c3=a9_Przywara?= Autocrypt: addr=andre.przywara@arm.com; prefer-encrypt=mutual; keydata= xsFNBFNPCKMBEAC+6GVcuP9ri8r+gg2fHZDedOmFRZPtcrMMF2Cx6KrTUT0YEISsqPoJTKld tPfEG0KnRL9CWvftyHseWTnU2Gi7hKNwhRkC0oBL5Er2hhNpoi8x4VcsxQ6bHG5/dA7ctvL6 kYvKAZw4X2Y3GTbAZIOLf+leNPiF9175S8pvqMPi0qu67RWZD5H/uT/TfLpvmmOlRzNiXMBm kGvewkBpL3R2clHquv7pB6KLoY3uvjFhZfEedqSqTwBVu/JVZZO7tvYCJPfyY5JG9+BjPmr+ REe2gS6w/4DJ4D8oMWKoY3r6ZpHx3YS2hWZFUYiCYovPxfj5+bOr78sg3JleEd0OB0yYtzTT esiNlQpCo0oOevwHR+jUiaZevM4xCyt23L2G+euzdRsUZcK/M6qYf41Dy6Afqa+PxgMEiDto ITEH3Dv+zfzwdeqCuNU0VOGrQZs/vrKOUmU/QDlYL7G8OIg5Ekheq4N+Ay+3EYCROXkstQnf YYxRn5F1oeVeqoh1LgGH7YN9H9LeIajwBD8OgiZDVsmb67DdF6EQtklH0ycBcVodG1zTCfqM AavYMfhldNMBg4vaLh0cJ/3ZXZNIyDlV372GmxSJJiidxDm7E1PkgdfCnHk+pD8YeITmSNyb 7qeU08Hqqh4ui8SSeUp7+yie9zBhJB5vVBJoO5D0MikZAODIDwARAQABzS1BbmRyZSBQcnp5 d2FyYSAoQVJNKSA8YW5kcmUucHJ6eXdhcmFAYXJtLmNvbT7CwXsEEwECACUCGwMGCwkIBwMC BhUIAgkKCwQWAgMBAh4BAheABQJTWSV8AhkBAAoJEAL1yD+ydue63REP/1tPqTo/f6StS00g NTUpjgVqxgsPWYWwSLkgkaUZn2z9Edv86BLpqTY8OBQZ19EUwfNehcnvR+Olw+7wxNnatyxo D2FG0paTia1SjxaJ8Nx3e85jy6l7N2AQrTCFCtFN9lp8Pc0LVBpSbjmP+Peh5Mi7gtCBNkpz KShEaJE25a/+rnIrIXzJHrsbC2GwcssAF3bd03iU41J1gMTalB6HCtQUwgqSsbG8MsR/IwHW XruOnVp0GQRJwlw07e9T3PKTLj3LWsAPe0LHm5W1Q+euoCLsZfYwr7phQ19HAxSCu8hzp43u zSw0+sEQsO+9wz2nGDgQCGepCcJR1lygVn2zwRTQKbq7Hjs+IWZ0gN2nDajScuR1RsxTE4WR lj0+Ne6VrAmPiW6QqRhliDO+e82riI75ywSWrJb9TQw0+UkIQ2DlNr0u0TwCUTcQNN6aKnru ouVt3qoRlcD5MuRhLH+ttAcmNITMg7GQ6RQajWrSKuKFrt6iuDbjgO2cnaTrLbNBBKPTG4oF D6kX8Zea0KvVBagBsaC1CDTDQQMxYBPDBSlqYCb/b2x7KHTvTAHUBSsBRL6MKz8wwruDodTM 4E4ToV9URl4aE/msBZ4GLTtEmUHBh4/AYwk6ACYByYKyx5r3PDG0iHnJ8bV0OeyQ9ujfgBBP B2t4oASNnIOeGEEcQ2rjzsFNBFNPCKMBEACm7Xqafb1Dp1nDl06aw/3O9ixWsGMv1Uhfd2B6 it6wh1HDCn9HpekgouR2HLMvdd3Y//GG89irEasjzENZPsK82PS0bvkxxIHRFm0pikF4ljIb 6tca2sxFr/H7CCtWYZjZzPgnOPtnagN0qVVyEM7L5f7KjGb1/o5EDkVR2SVSSjrlmNdTL2Rd zaPqrBoxuR/y/n856deWqS1ZssOpqwKhxT1IVlF6S47CjFJ3+fiHNjkljLfxzDyQXwXCNoZn BKcW9PvAMf6W1DGASoXtsMg4HHzZ5fW+vnjzvWiC4pXrcP7Ivfxx5pB+nGiOfOY+/VSUlW/9 GdzPlOIc1bGyKc6tGREH5lErmeoJZ5k7E9cMJx+xzuDItvnZbf6RuH5fg3QsljQy8jLlr4S6 8YwxlObySJ5K+suPRzZOG2+kq77RJVqAgZXp3Zdvdaov4a5J3H8pxzjj0yZ2JZlndM4X7Msr P5tfxy1WvV4Km6QeFAsjcF5gM+wWl+mf2qrlp3dRwniG1vkLsnQugQ4oNUrx0ahwOSm9p6kM CIiTITo+W7O9KEE9XCb4vV0ejmLlgdDV8ASVUekeTJkmRIBnz0fa4pa1vbtZoi6/LlIdAEEt PY6p3hgkLLtr2GRodOW/Y3vPRd9+rJHq/tLIfwc58ZhQKmRcgrhtlnuTGTmyUqGSiMNfpwAR AQABwsFfBBgBAgAJBQJTTwijAhsMAAoJEAL1yD+ydue64BgP/33QKczgAvSdj9XTC14wZCGE U8ygZwkkyNf021iNMj+o0dpLU48PIhHIMTXlM2aiiZlPWgKVlDRjlYuc9EZqGgbOOuR/pNYA JX9vaqszyE34JzXBL9DBKUuAui8z8GcxRcz49/xtzzP0kH3OQbBIqZWuMRxKEpRptRT0wzBL O31ygf4FRxs68jvPCuZjTGKELIo656/Hmk17cmjoBAJK7JHfqdGkDXk5tneeHCkB411p9WJU vMO2EqsHjobjuFm89hI0pSxlUoiTL0Nuk9Edemjw70W4anGNyaQtBq+qu1RdjUPBvoJec7y/ EXJtoGxq9Y+tmm22xwApSiIOyMwUi9A1iLjQLmngLeUdsHyrEWTbEYHd2sAM2sqKoZRyBDSv ejRvZD6zwkY/9nRqXt02H1quVOP42xlkwOQU6gxm93o/bxd7S5tEA359Sli5gZRaucpNQkwd KLQdCvFdksD270r4jU/rwR2R/Ubi+txfy0dk2wGBjl1xpSf0Lbl/KMR5TQntELfLR4etizLq Xpd2byn96Ivi8C8u9zJruXTueHH8vt7gJ1oax3yKRGU5o2eipCRiKZ0s/T7fvkdq+8beg9ku fDO4SAgJMIl6H5awliCY2zQvLHysS/Wb8QuB09hmhLZ4AifdHyF1J5qeePEhgTA+BaUbiUZf i4aIXCH3Wv6K Organization: ARM Ltd. Message-ID: <87569683-509e-96e6-17f9-c1734a8b32d4@arm.com> Date: Mon, 11 May 2020 12:44:26 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.7.0 MIME-Version: 1.0 In-Reply-To: <20200511112014.GA3322@Mani-XPS-13-9360> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: dmaengine-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: dmaengine@vger.kernel.org On 11/05/2020 12:20, Manivannan Sadhasivam wrote: Hi, > On Mon, May 11, 2020 at 04:15:57PM +0530, Amit Tomer wrote: >> Hi >> >> Thanks for the reply. >> >>> I'm in favor of getting rid of bitfields due to its not so defined way of >>> working (and forgive me for using it in first place) but I don't quite like >>> the current approach. >> >> Because , its less readable the way we are writing to those different fields ? >> But this can be made more verbose by adding some comments around . >> > > I don't like the way the hw linked lists are accessed (using an array with > enums). But honestly this is the most sane way of doing this, see below. >>> Rather I'd like to have custom bitmasks (S900/S700/S500?) for writing to those >>> fields. >>> >> I think S900 and S500 are same as pointed out by Cristian. and I didn't get by >> creating custom bitmasks for it ? >> >> Did you mean function like: >> >> lli->hw[OWL_DMADESC_FLEN]= llc_hw_FLEN(len, FCNT_VALUE, FCNT_SHIFT); >> > > I meant to keep using old struct for accessing the linked list and replacing > bitfields with masks as below: > > struct owl_dma_lli_hw { > ... > u32 flen; > u32 fcnt; > ... > }; And is think this is the wrong way of modelling hardware defined register fields. C structs have no guarantee of not introducing padding in between fields, the only guarantee you get is that the first member has no padding *before* it: C standard, section 6.7.2.1, end of paragraph 15: "There may be unnamed padding within a structure object, but not at its beginning." Arrays in C on the contrary have very much this guarantee: The members are next to each other, no padding. I see that structs are sometimes used in this function, but it's much less common in the kernel than in other projects (U-Boot comes to mind). It typically works, because common compiler *implementations* provide this guarantee, but we should not rely on this. So: Using enums for the keys provides a natural way of increasing indices, without gaps. Also you get this nice and automatic size value by making this the last member of the enum. Arrays provide the guarantee of consecutive allocation. We can surely have a look at the masking problem, but this would need to be runtime determined masks, which tend to become "wordy". There can be simplifications, for instance I couldn't find where the frame length is really limited for the S900 (it must be less than 1MB). Since the S700 supports *more* than that, there is no need to limit this differently. Cheers, Andre. > > hw->flen = len & OWL_S900_DMA_FLEN_MASK; > hw->fcnt = 1 & OWL_S900_DMA_FCNT_MASK; > > Then you can use different masks for S700/S900 based on the compatible. > > Thanks, > Mani > >> Thanks >> -Amit