From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dpdk.org (dpdk.org [92.243.14.124]) by inbox.dpdk.org (Postfix) with ESMTP id 20FFCA0513; Wed, 15 Jan 2020 15:36:30 +0100 (CET) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 62F191C0D8; Wed, 15 Jan 2020 15:36:29 +0100 (CET) Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by dpdk.org (Postfix) with ESMTP id 3C0841C013 for ; Wed, 15 Jan 2020 15:36:28 +0100 (CET) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga007.fm.intel.com ([10.253.24.52]) by fmsmga101.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 15 Jan 2020 06:36:27 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,322,1574150400"; d="scan'208";a="219294098" Received: from fyigit-mobl.ger.corp.intel.com (HELO [10.237.221.35]) ([10.237.221.35]) by fmsmga007.fm.intel.com with ESMTP; 15 Jan 2020 06:36:24 -0800 To: "Iremonger, Bernard" , Andrew Rybchenko , "dev@dpdk.org" , "Xing, Beilei" , "Zhang, Qi Z" , "Doherty, Declan" Cc: "Ananyev, Konstantin" , "Byrne, Stephen1" , "Zhang, Helin" , Ori Kam , Thomas Monjalon , "Richardson, Bruce" , "Kinsella, Ray" , David Marchand , Luca Boccassi , Jerin Jacob , Olivier MATZ References: <1578669630-23866-1-git-send-email-bernard.iremonger@intel.com> <1579010128-15794-13-git-send-email-bernard.iremonger@intel.com> <43ef6a41-6114-8a0c-bc64-3933d4dc9f7d@solarflare.com> <651f4fc5-ccee-a0fd-096e-d67c25248352@intel.com> <6b5811ba-2c98-6335-a723-87f64b137b0c@solarflare.com> From: Ferruh Yigit Autocrypt: addr=ferruh.yigit@intel.com; prefer-encrypt=mutual; keydata= mQINBFXZCFABEADCujshBOAaqPZpwShdkzkyGpJ15lmxiSr3jVMqOtQS/sB3FYLT0/d3+bvy qbL9YnlbPyRvZfnP3pXiKwkRoR1RJwEo2BOf6hxdzTmLRtGtwWzI9MwrUPj6n/ldiD58VAGQ +iR1I/z9UBUN/ZMksElA2D7Jgg7vZ78iKwNnd+vLBD6I61kVrZ45Vjo3r+pPOByUBXOUlxp9 GWEKKIrJ4eogqkVNSixN16VYK7xR+5OUkBYUO+sE6etSxCr7BahMPKxH+XPlZZjKrxciaWQb +dElz3Ab4Opl+ZT/bK2huX+W+NJBEBVzjTkhjSTjcyRdxvS1gwWRuXqAml/sh+KQjPV1PPHF YK5LcqLkle+OKTCa82OvUb7cr+ALxATIZXQkgmn+zFT8UzSS3aiBBohg3BtbTIWy51jNlYdy ezUZ4UxKSsFuUTPt+JjHQBvF7WKbmNGS3fCid5Iag4tWOfZoqiCNzxApkVugltxoc6rG2TyX CmI2rP0mQ0GOsGXA3+3c1MCdQFzdIn/5tLBZyKy4F54UFo35eOX8/g7OaE+xrgY/4bZjpxC1 1pd66AAtKb3aNXpHvIfkVV6NYloo52H+FUE5ZDPNCGD0/btFGPWmWRmkPybzColTy7fmPaGz cBcEEqHK4T0aY4UJmE7Ylvg255Kz7s6wGZe6IR3N0cKNv++O7QARAQABtCVGZXJydWggWWln aXQgPGZlcnJ1aC55aWdpdEBpbnRlbC5jb20+iQJUBBMBCgA+AhsDAh4BAheABQsJCAcDBRUK CQgLBRYCAwEAFiEE0jZTh0IuwoTjmYHH+TPrQ98TYR8FAl1meboFCQlupOoACgkQ+TPrQ98T YR9ACBAAv2tomhyxY0Tp9Up7mNGLfEdBu/7joB/vIdqMRv63ojkwr9orQq5V16V/25+JEAD0 60cKodBDM6HdUvqLHatS8fooWRueSXHKYwJ3vxyB2tWDyZrLzLI1jxEvunGodoIzUOtum0Ce gPynnfQCelXBja0BwLXJMplM6TY1wXX22ap0ZViC0m714U5U4LQpzjabtFtjT8qOUR6L7hfy YQ72PBuktGb00UR/N5UrR6GqB0x4W41aZBHXfUQnvWIMmmCrRUJX36hOTYBzh+x86ULgg7H2 1499tA4o6rvE13FiGccplBNWCAIroAe/G11rdoN5NBgYVXu++38gTa/MBmIt6zRi6ch15oLA Ln2vHOdqhrgDuxjhMpG2bpNE36DG/V9WWyWdIRlz3NYPCDM/S3anbHlhjStXHOz1uHOnerXM 1jEjcsvmj1vSyYoQMyRcRJmBZLrekvgZeh7nJzbPHxtth8M7AoqiZ/o/BpYU+0xZ+J5/szWZ aYxxmIRu5ejFf+Wn9s5eXNHmyqxBidpCWvcbKYDBnkw2+Y9E5YTpL0mS0dCCOlrO7gca27ux ybtbj84aaW1g0CfIlUnOtHgMCmz6zPXThb+A8H8j3O6qmPoVqT3qnq3Uhy6GOoH8Fdu2Vchh TWiF5yo+pvUagQP6LpslffufSnu+RKAagkj7/RSuZV25Ag0EV9ZMvgEQAKc0Db17xNqtSwEv mfp4tkddwW9XA0tWWKtY4KUdd/jijYqc3fDD54ESYpV8QWj0xK4YM0dLxnDU2IYxjEshSB1T qAatVWz9WtBYvzalsyTqMKP3w34FciuL7orXP4AibPtrHuIXWQOBECcVZTTOdZYGAzaYzxiA ONzF9eTiwIqe9/oaOjTwTLnOarHt16QApTYQSnxDUQljeNvKYt1lZE/gAUUxNLWsYyTT+22/ vU0GDUahsJxs1+f1yEr+OGrFiEAmqrzpF0lCS3f/3HVTU6rS9cK3glVUeaTF4+1SK5ZNO35p iVQCwphmxa+dwTG/DvvHYCtgOZorTJ+OHfvCnSVjsM4kcXGjJPy3JZmUtyL9UxEbYlrffGPQ I3gLXIGD5AN5XdAXFCjjaID/KR1c9RHd7Oaw0Pdcq9UtMLgM1vdX8RlDuMGPrj5sQrRVbgYH fVU/TQCk1C9KhzOwg4Ap2T3tE1umY/DqrXQgsgH71PXFucVjOyHMYXXugLT8YQ0gcBPHy9mZ qw5mgOI5lCl6d4uCcUT0l/OEtPG/rA1lxz8ctdFBVOQOxCvwRG2QCgcJ/UTn5vlivul+cThi 6ERPvjqjblLncQtRg8izj2qgmwQkvfj+h7Ex88bI8iWtu5+I3K3LmNz/UxHBSWEmUnkg4fJl Rr7oItHsZ0ia6wWQ8lQnABEBAAGJAjwEGAEKACYCGwwWIQTSNlOHQi7ChOOZgcf5M+tD3xNh HwUCXWZ5wAUJB3FgggAKCRD5M+tD3xNhH2O+D/9OEz62YuJQLuIuOfL67eFTIB5/1+0j8Tsu o2psca1PUQ61SZJZOMl6VwNxpdvEaolVdrpnSxUF31kPEvR0Igy8HysQ11pj8AcgH0a9FrvU /8k2Roccd2ZIdpNLkirGFZR7LtRw41Kt1Jg+lafI0efkiHKMT/6D/P1EUp1RxOBNtWGV2hrd 0Yg9ds+VMphHHU69fDH02SwgpvXwG8Qm14Zi5WQ66R4CtTkHuYtA63sS17vMl8fDuTCtvfPF HzvdJLIhDYN3Mm1oMjKLlq4PUdYh68Fiwm+boJoBUFGuregJFlO3hM7uHBDhSEnXQr5mqpPM 6R/7Q5BjAxrwVBisH0yQGjsWlnysRWNfExAE2sRePSl0or9q19ddkRYltl6X4FDUXy2DTXa9 a+Fw4e1EvmcF3PjmTYs9IE3Vc64CRQXkhujcN4ZZh5lvOpU8WgyDxFq7bavFnSS6kx7Tk29/ wNJBp+cf9qsQxLbqhW5kfORuZGecus0TLcmpZEFKKjTJBK9gELRBB/zoN3j41hlEl7uTUXTI JQFLhpsFlEdKLujyvT/aCwP3XWT+B2uZDKrMAElF6ltpTxI53JYi22WO7NH7MR16Fhi4R6vh FHNBOkiAhUpoXRZXaCR6+X4qwA8CwHGqHRBfYFSU/Ulq1ZLR+S3hNj2mbnSx0lBs1eEqe2vh cA== Message-ID: <1bec7eef-3f01-5629-22ab-83cfd46a1e07@intel.com> Date: Wed, 15 Jan 2020 14:36:23 +0000 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Subject: Re: [dpdk-dev] [PATCH v5 12/14] librte_ethdev: add ESP and AH flow types to RSS X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" On 1/15/2020 2:11 PM, Iremonger, Bernard wrote: > Hi Ferruh, > > > >> Subject: Re: [dpdk-dev] [PATCH v5 12/14] librte_ethdev: add ESP and AH >> flow types to RSS >> >> On 1/15/2020 10:55 AM, Andrew Rybchenko wrote: >>> On 1/15/20 1:44 PM, Ferruh Yigit wrote: >>>> On 1/15/2020 9:13 AM, Andrew Rybchenko wrote: >>>>> On 1/14/20 9:45 PM, Ferruh Yigit wrote: >>>>>> On 1/14/2020 1:55 PM, Bernard Iremonger wrote: >>>>>>> Add macros for the following protocols in the DDP esp-ah profile: >>>>>>> ESP >>>>>>> AH >>>>>>> >>>>>>> Add the following RSS macro for IPsec: >>>>>>> ETH_RSS_IPSEC >>>>>>> >>>>>>> Signed-off-by: Bernard Iremonger >>>>>> +Ori and other ethdev maintainers. >>>>>> >>>>>> Ori, can you please check this patch? >>>>>> >>>>>>> --- >>>>>>> lib/librte_ethdev/rte_ethdev.h | 14 +++++++++++++- >>>>>>> 1 file changed, 13 insertions(+), 1 deletion(-) >>>>>>> >>>>>>> diff --git a/lib/librte_ethdev/rte_ethdev.h >>>>>>> b/lib/librte_ethdev/rte_ethdev.h index 18a9def..208ec90 100644 >>>>>>> --- a/lib/librte_ethdev/rte_ethdev.h >>>>>>> +++ b/lib/librte_ethdev/rte_ethdev.h >>>>>>> @@ -484,7 +484,9 @@ struct rte_eth_rss_conf { >>>>>>> #define RTE_ETH_FLOW_NVGRE 21 /**< NVGRE protocol >> based flow */ >>>>>>> #define RTE_ETH_FLOW_VXLAN_GPE 22 /**< VXLAN-GPE >> protocol based flow */ >>>>>>> #define RTE_ETH_FLOW_GTPU 23 /**< GTPU protocol based >> flow */ >>>>>>> -#define RTE_ETH_FLOW_MAX 24 >>>>>>> +#define RTE_ETH_FLOW_AH 24 /**< AH protocol based flow >> */ >>>>>>> +#define RTE_ETH_FLOW_ESP 25 /**< ESP protocol based >> flow */ >>>>>>> +#define RTE_ETH_FLOW_MAX 26 >>>>> Isn't changing RTE_ETH_FLOW_MAX value breaking ABI? >>>>> Is v20.11 target release of the patch? >>>> I can't see how this can cause an ABI break, unless the >>>> 'RTE_ETH_FLOW_MAX' value used as size of an array in the middle of a >> struct. >>>> There is 'struct rte_eth_fdir_flex_conf' but the array is at the end >>>> there, so it should be OK, unless that struct is not in the middle of another >> struct. >>> >>> rte_eth_fdir_flex_conf -> rte_fdir_conf -> rte_eth_conf (in the >>> middle) >> >> Yes, this looks like an ABI break and this is very annoying not able to even add >> a new RTE_FLOW type. >> >> We need to find a proper way to handle this, at first glance I can see stop >> using _MAX macros for the array size can work and perhaps we can use >> another big enough hardcoded value for all similar array size. Any other >> option? >> >> But we can do this on 20.11, we need a solution until that time. > > This patch is required to handle RSS for the esp-ah.pkg DDP profile, it does not affect the i40e FD and testpmd changes in this patch set. > The esp-ah.pkg DDP profile is not released yet. > > Given that the merge deadline is today, I propose to drop this patch from the v6 patch set. +1, if it can be separated, better to do it to not block rest of the work. > >> >>> >>>> And there are values calculated from 'RTE_ETH_FLOW_MAX', like >>>> 'RTE_FLOW_MASK_ARRAY_SIZE', same concern applies there, it very hard >> to follow. >>>> >>>> Bernard, >>>> >>>> Can you please run the ABI version script to be sure this is not breaking >> the ABI? >>>> >>>> >>>>>>> >>>>>>> /* >>>>>>> * Below macros are defined for RSS offload types, they can be >>>>>>> used to @@ -511,6 +513,12 @@ struct rte_eth_rss_conf { >>>>>>> #define ETH_RSS_GENEVE (1ULL << 20) >>>>>>> #define ETH_RSS_NVGRE (1ULL << 21) >>>>>>> #define ETH_RSS_GTPU (1ULL << 23) >>>>>>> +#define ETH_RSS_AH (1ULL << 24) >>>>>>> +#define ETH_RSS_ESP (1ULL << 25) >>>>>>> + >>>>>>> + >>>>>>> + >>>>>>> + >>>>>>> >>>>>>> /* >>>>>>> * We use the following macros to combine with above ETH_RSS_* >>>>>>> for @@ -571,6 +579,10 @@ rte_eth_rss_hf_refine(uint64_t rss_hf) >>>>>>> ETH_RSS_NONFRAG_IPV4_SCTP | \ >>>>>>> ETH_RSS_NONFRAG_IPV6_SCTP) >>>>>>> >>>>>>> +#define ETH_RSS_IPSEC ( \ >>>>>>> +ETH_RSS_AH | \ >>>>>>> +ETH_RSS_ESP) >>>>>>> + >>>>>>> #define ETH_RSS_TUNNEL ( \ >>>>>>> ETH_RSS_VXLAN | \ >>>>>>> ETH_RSS_GENEVE | \ >>>>>>> >>> > Regards, > > Bernard. >