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 ECAA9A0471 for ; Mon, 9 Sep 2019 13:23:42 +0200 (CEST) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 38E471EAB5; Mon, 9 Sep 2019 13:23:42 +0200 (CEST) Received: from mga04.intel.com (mga04.intel.com [192.55.52.120]) by dpdk.org (Postfix) with ESMTP id 7C07A2C4F for ; Mon, 9 Sep 2019 13:23:40 +0200 (CEST) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga005.jf.intel.com ([10.7.209.41]) by fmsmga104.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 09 Sep 2019 04:23:39 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.64,484,1559545200"; d="scan'208";a="359460023" Received: from fyigit-mobl.ger.corp.intel.com (HELO [10.237.221.4]) ([10.237.221.4]) by orsmga005.jf.intel.com with ESMTP; 09 Sep 2019 04:23:37 -0700 To: "Wang, Haiyue" , Ray Kinsella Cc: "dev@dpdk.org" , "Sun, Chenmin" References: <1565665572-65495-1-git-send-email-haiyue.wang@intel.com> <1565665572-65495-2-git-send-email-haiyue.wang@intel.com> <20190812202436.662e6ca8@hermes.lan> From: Ferruh Yigit Openpgp: preference=signencrypt 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: <75f88e4b-97ca-7e1f-1385-8e69f659be7f@intel.com> Date: Mon, 9 Sep 2019 12:23:36 +0100 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit Subject: Re: [dpdk-dev] [RFC v2 1/3] ethdev: add the API for getting trace information 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 9/7/2019 3:42 AM, Wang, Haiyue wrote: >> -----Original Message----- >> From: Yigit, Ferruh >> Sent: Friday, September 6, 2019 22:22 >> To: Ray Kinsella ; Wang, Haiyue >> Cc: dev@dpdk.org >> Subject: Re: [dpdk-dev] [RFC v2 1/3] ethdev: add the API for getting trace information >> >> On 8/13/2019 1:51 PM, Ray Kinsella wrote: >>> >>> >>> On 13/08/2019 04:24, Stephen Hemminger wrote: >>>> On Tue, 13 Aug 2019 11:06:10 +0800 >>>> Haiyue Wang wrote: >>>> >>>>> Enhance the PMD to support retrieving trace information like >>>>> Rx/Tx burst selection etc. >>>>> >>>>> Signed-off-by: Haiyue Wang >>>>> --- >>>>> lib/librte_ethdev/rte_ethdev.c | 18 ++++++++++++++++++ >>>>> lib/librte_ethdev/rte_ethdev.h | 9 +++++++++ >>>>> lib/librte_ethdev/rte_ethdev_core.h | 4 ++++ >>>>> 3 files changed, 31 insertions(+) >>>>> >>>>> diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c >>>>> index 17d183e..6098fad 100644 >>>>> --- a/lib/librte_ethdev/rte_ethdev.c >>>>> +++ b/lib/librte_ethdev/rte_ethdev.c >>>>> @@ -4083,6 +4083,24 @@ rte_eth_tx_queue_info_get(uint16_t port_id, uint16_t queue_id, >>>>> } >>>>> >>>>> int >>>>> +rte_eth_trace_info_get(uint16_t port_id, uint16_t queue_id, >>>>> + enum rte_eth_trace type, char *buf, int sz) >> >> Better to use struct as argument instead of individual variables because it is >> easier to extend the struct later if needed. >> >>>>> +{ >>>>> + struct rte_eth_dev *dev; >>>>> + >>>>> + RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV); >>>>> + >>>>> + if (buf == NULL) >>>>> + return -EINVAL; >>>>> + >>>>> + dev = &rte_eth_devices[port_id]; >>>>> + >>>>> + RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->trace_info_get, -ENOTSUP); >>>>> + >>>>> + return dev->dev_ops->trace_info_get(dev, queue_id, type, buf, sz); >>>> >>>> What if queueid is out of bounds? >>>> >>>> The bigger problem is that this information is like a log message >>>> and unstructured, which makes it device specific and useless for automation. >>> >>> IMHO - this is much better implemented as a capability bitfield, that >>> can be queried. >> >> +1 to return the datapath capability as bitfield. >> >> Also +1 to have a new API, >> - I am not sure about the API name, 'rte_eth_trace_info_get()', can we find >> something better instead of 'trace' there. >> - I think we should limit this API only to get current datapath configuration, >> for clarity of the API don't return capability or not datapath related config. >> >> Also this information not always supported in queue level, what do you think >> having ability to get this information in port level, >> like this API can return a struct, which may have a field that says if the >> output is for queue or port, or this can be another bitfield, what do you think? >> > > #define RX_SCALAR (1ULL < 0) > #define RX_VECTOR_AVX2 ... What about having RX_VECTOR value, later another bit group for the details of the vectorization: SSE AVX2 AVX512 NEON ALTIVEC Since above options can exist together, what about using values for them instead of bitfields? Reserving 4 bits, 2^4 = 16, can be enough I think for long term. > ... > #define RX_SCATTER ... > #define RX_BULK_ALLOC > #define TX_SCALAR > #define TX_VECTOR_AVX2 .. > ... > #define TX_SIMPLE And what about implementing these as 'enum' instead of defines? > > struct rte_pkt_burst_info { > bool per_queue_support; /* Per queue has each rx/tx burst setting */ > uint64_t burst_infos; 'infos' can be wrong linguistically, I think better to stick to 'info' > }; > > int > rte_eth_pkt_burst_info_get(uint16_t port_id, bool is_rx, uint16_t queue_id, > struct rte_pkt_burst_info *pbinfo) Naming is the hard ;) I am not sure about 'burst_info' as well, I think they are exactly "burst mode function types" that recv/send burst of packets. It feels like it will provide information about all the burst mode related things like burst size etc.. but this is only for the function type. And suggestion is welcome :) > > Rx/Tx shares the 64 bits definition, but return according to 'is_rx'. Why 'is_rx' required, is it because 'per_queue_support' field? If so I think better to add Rx and Tx version of it and drop the 'is_rx' from the API. > Application can call with 'queue_id = 0' firstly, if the Rx/Tx queue > support queue level burst setting, then 'per_queue_support = true', then > it can iterate to get the burst info with different 'queue_id', of course, > the 'per_queue_support = true' will be returned always. I am not sure if that is what you meant but there shouldn't be any special behavior for "queue_id = 0", for the devices that doesn't support queue level, they can send same value for each queue with "per_queue_support = false" An application iterating the over queues can use same API without worrying the 'per_queue_support' value, for the application that assumes the burst mode function should be same for the port may want to verify this in advance by checking the 'per_queue_support' value. > > How about this ? > > >>> >>>> >>>> Why not just keep it in the log like it is now? >>>> >>>>> int rte_eth_tx_queue_info_get(uint16_t port_id, uint16_t queue_id, >>>>> struct rte_eth_txq_info *qinfo); >>>>> >>>>> +int >>>>> +rte_eth_trace_info_get(uint16_t port_id, uint16_t queue_id, >>>>> + enum rte_eth_trace type, char *buf, int sz); >>>>> + >>>> >>>> You didn't run checkpatch, otherwise you would have seen complaints >>>> about not listing API as experimental. >>>> >>>> Also the API would have to be in the map file as well. >>>> >>>> Docbook comments are also missing. >>>> >>>> >>>> >>>> >