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 8C080A04B0; Thu, 10 Dec 2020 10:24:05 +0100 (CET) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id C7F6BBC8A; Thu, 10 Dec 2020 10:24:03 +0100 (CET) Received: from mga17.intel.com (mga17.intel.com [192.55.52.151]) by dpdk.org (Postfix) with ESMTP id 53F49BC88 for ; Thu, 10 Dec 2020 10:24:01 +0100 (CET) IronPort-SDR: LUgvARJHjboi/ARY9/OZ1Gqlo/DZxGYRq4nDrvnyX2X2R7XioyeW5fhfePOTPLSuEyfjxgADpi rnvwXp+BtHpA== X-IronPort-AV: E=McAfee;i="6000,8403,9830"; a="154031848" X-IronPort-AV: E=Sophos;i="5.78,408,1599548400"; d="scan'208";a="154031848" Received: from orsmga008.jf.intel.com ([10.7.209.65]) by fmsmga107.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 10 Dec 2020 01:23:59 -0800 IronPort-SDR: 7UY28Wkwkok4TWNUVDIzJAln5aShoPetKH2mWlyaj6OiTqvcUj+kSiEtsY8/fQ154NNzeG5Ybz fJyssDworJ3Q== X-IronPort-AV: E=Sophos;i="5.78,408,1599548400"; d="scan'208";a="364547687" Received: from fyigit-mobl1.ger.corp.intel.com (HELO [10.213.200.38]) ([10.213.200.38]) by orsmga008-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 10 Dec 2020 01:23:57 -0800 To: Andrew Boyer Cc: dev@dpdk.org, Alfredo Cardigliano , Thomas Monjalon , David Marchand References: <20201203203418.15064-1-aboyer@pensando.io> <20201204201646.51746-4-aboyer@pensando.io> <8c32bc91-6583-cd3d-a1e6-979bda122eb3@intel.com> <50603b0a-07a7-b413-b4c5-1fd535befd3f@intel.com> <55dfbcec-4974-da24-8a46-adbbc57850b4@intel.com> <6266CE88-1FCD-471D-8390-33BA77707BD7@pensando.io> From: Ferruh Yigit Message-ID: <63ce042d-a7f7-2770-8a1e-0c04bd894f1d@intel.com> Date: Thu, 10 Dec 2020 09:23:54 +0000 MIME-Version: 1.0 In-Reply-To: <6266CE88-1FCD-471D-8390-33BA77707BD7@pensando.io> Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: 8bit Subject: Re: [dpdk-dev] [PATCH v3 3/9] net/ionic: update documentation and MAINTAINERS 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 12/9/2020 7:05 PM, Andrew Boyer wrote: > > >> On Dec 9, 2020, at 12:15 PM, Ferruh Yigit wrote: >> >> On 12/9/2020 4:24 PM, Andrew Boyer wrote: >>>> On Dec 9, 2020, at 10:24 AM, Ferruh Yigit > wrote: >>>> >>>> On 12/9/2020 2:36 PM, Andrew Boyer wrote: >>>>> Please respond to my questions this time. I have ~70 more patches to post by December 20. >>>>>> On Dec 9, 2020, at 7:03 AM, Ferruh Yigit >> wrote: >>>>>> >>>>>> On 12/4/2020 8:16 PM, Andrew Boyer wrote: >>>>>>> The UNMAINTAINED flag will be removed in a future patch. >>>>>>> Signed-off-by: Andrew Boyer >> >>>>>>> --- >>>>>>> MAINTAINERS | 3 ++- >>>>>>> doc/guides/nics/features/ionic.ini | 2 ++ >>>>>>> doc/guides/nics/ionic.rst | 13 +++++++------ >>>>>>> 3 files changed, 11 insertions(+), 7 deletions(-) >>>>>>> diff --git a/MAINTAINERS b/MAINTAINERS >>>>>>> index eafe9f8c4..6534983c1 100644 >>>>>>> --- a/MAINTAINERS >>>>>>> +++ b/MAINTAINERS >>>>>>> @@ -841,7 +841,8 @@ F: drivers/net/pfe/ >>>>>>> F: doc/guides/nics/features/pfe.ini >>>>>>> Pensando ionic - UNMAINTAINED >>>>>>> -M: Alfredo Cardigliano >> >>>>>>> +M: Andrew Boyer >> >>>>>>> +M: Pensando Drivers >> >>>>>> >>>>>> Same comment from previous version, please don't add group as maintainer, only actual people. >>>>> I responded to your original comment about this back in November. Is there an official DPDK policy against doing this? Is it your preference? We would very much prefer to have this in the file as a fallback. As long as there is still at least one person listed, what is the harm? >>>> >>>> There is no official policy against it as far as I know. >>>> >>>> The problem with the groups is we don't know who is behind it, it blurs who is the owner/responsible of the component. Actual people makes it clear that who is responsible. >>>> >>>> Why do you prefer to add a group as maintainer? >>> Because if I am on leave for some reason, one of the other handful of maintainers might be able to help someone with a problem or a question. >>> If I am listed specifically, doesn’t that make clear “who is the owner/responsible” for ionic PMD? What harm does having drivers@ listed do? >> >> If you want backups, pick and list a few of those handful maintainers and add to the list, so we can know who they are. >> >> As already said problem is we don't know who is behind a group, if they are reliable or not, or is there really someone or not, what happens if group become silent? >> You can cc that group in your patches, that is commonly done, so they can be part of the development process, but the maintainers file is to define responsible people, adding a group is hiding actual responsible people. > > Fine. Perhaps you could document somewhere that this requirement exists to save future people the trouble. > > >>>>>>> F: drivers/net/ionic/ >>>>>>> F: doc/guides/nics/ionic.rst >>>>>>> F: doc/guides/nics/features/ionic.ini >>>>>>> diff --git a/doc/guides/nics/features/ionic.ini b/doc/guides/nics/features/ionic.ini >>>>>>> index 083c7bd99..dd29dbed6 100644 >>>>>>> --- a/doc/guides/nics/features/ionic.ini >>>>>>> +++ b/doc/guides/nics/features/ionic.ini >>>>>>> @@ -8,6 +8,7 @@ Speed capabilities = Y >>>>>>> Link status = Y >>>>>>> Link status event = Y >>>>>>> Queue start/stop = Y >>>>>>> +Lock-free Tx queue = Y >>>>>> >>>>>> Are you sure this is supported? >>>>>> Since it is not advertised as capability, I think this can't be claimed as supported, but still even after this is added as capability, can you please confirm your device supports multiple core enqueue to same queue without locks? >>>>> I misunderstood the meaning of this flag, will remove. >>>>>>> MTU update = Y >>>>>>> Jumbo frame = Y >>>>>>> Scattered Rx = Y >>>>>>> @@ -19,6 +20,7 @@ Unicast MAC filter = Y >>>>>>> RSS hash = Y >>>>>>> RSS key update = Y >>>>>>> RSS reta update = Y >>>>>>> +SR-IOV = Y >>>>>> >>>>>> Can you please explain what is exactly supported? Like can DPDK drive both PF & VF? >>>>> Yes. The PMD does not distinguish between PFs and VFs. >>>>>> >>>>>> <...> >>>>>> >>>>>>> @@ -7,15 +7,16 @@ IONIC Driver >>>>>>> The ionic driver provides support for Pensando server adapters. >>>>>>> It currently supports the below models: >>>>>>> -- `Naples DSC-25 >>`_ >>>>>>> -- `Naples DSC-100 >>`_ >>>>>>> +- DSC-25 dual-port 25G Distributed Services Card >>>>>>> +- DSC-100 dual-port 100G Distributed Services Card >>>>>>> >>>>>> >>>>>> Same comment from previous version, can you provide link for these devices, it is hard to find the devices from the main site. >>>>> And my same response from your previous comment. I do not control the website and do not wish to put stale PDF links in this document, which will live forever. The text includes the URL of the page containing links to the PDFs. Why is this not acceptable? >>>> >>>> The request is to put links to the products that you are providing the driver for. This is to help people that are already interested your driver and reading your driver document, to reach to the product information easily. >>>> >>>> The request is NOT to provide pdf etc, just a reference to the product. Don't you advertise your product in your official web site? If your product information is not visible/hidden, why you are providing the open source drivers for it? >>> Does this line in the doc not satisfy your request? >>> +The `Documents >`_ page contains Product Briefs and other product information. >> >> No it doesn't, it is not clear which document/product you are referring to. >> Why it is hard to provide the link of products that your driver is for? >> > > Fine. These links will become stale next time the files are updated, just like the links in the 20.02-20.11 releases are stale. > > https://pensando.io/wp-content/uploads/2020/03/Pensando-DSC-25-Product-Brief.pdf > https://pensando.io/wp-content/uploads/2020/03/Pensando-DSC-100-Product-Brief.pdf > If you are sure that they will go stale, that is also not good solution. Why they are keep getting stale? Can't you communicate internally in your company for permanent link?