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 7F648A0C41 for ; Thu, 16 Sep 2021 10:09:04 +0200 (CEST) Received: from [217.70.189.124] (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 6A100410E8; Thu, 16 Sep 2021 10:09:04 +0200 (CEST) Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by mails.dpdk.org (Postfix) with ESMTP id DB78240E78 for ; Thu, 16 Sep 2021 10:09:02 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1631779742; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=7WhHhmmU9kapA3Qt/QWCVvFOdsAaLJGSolN/1IiQ0wI=; b=KiDPTo6RruOQ1FuoPtDtF2N7WRxgc5FIGx1bmmxJ7Ns0V1fNvKdWr0AnMUx6rMWMJ/X1iH zrDl2550iqRMbDYkj2LA8kGrv0U3nfLjtNmuks/tNScT1c/SGFjmI3R+qjjwEsiLDenobJ q6L4STX5ViSTyMYnuFrNhwDu0ZJ0Qec= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-434--2eY04fqPPO1QOAPNCQcqg-1; Thu, 16 Sep 2021 04:08:59 -0400 X-MC-Unique: -2eY04fqPPO1QOAPNCQcqg-1 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.13]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id A132480124F; Thu, 16 Sep 2021 08:08:57 +0000 (UTC) Received: from [10.39.208.33] (unknown [10.39.208.33]) by smtp.corp.redhat.com (Postfix) with ESMTPS id 3EB841B5C2; Thu, 16 Sep 2021 08:08:51 +0000 (UTC) To: "Li, Xiaoyun" , =?UTF-8?Q?N=c3=a9lio_Laranjeiro?= , "Yigit, Ferruh" Cc: Andrew Rybchenko , "dev@dpdk.org" , "Xia, Chenbo" , "amorenoz@redhat.com" , "david.marchand@redhat.com" , "michaelba@nvidia.com" , "viacheslavo@nvidia.com" , "stable@dpdk.org" , Shahaf Shuler References: <20210910091734.7023-1-maxime.coquelin@redhat.com> <20210910091734.7023-3-maxime.coquelin@redhat.com> <7548b2d2-fd23-4a1a-59d7-b3424912afcc@redhat.com> <85036fee-62f0-5d61-b451-65ee428ee2bc@oktetlabs.ru> <1daeb2f2-2122-c546-3eb4-f0d7fba035f0@redhat.com> From: Maxime Coquelin Message-ID: <0a1866ff-3776-9edf-055e-d20efa6d6ec7@redhat.com> Date: Thu, 16 Sep 2021 10:08:49 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.11.0 MIME-Version: 1.0 In-Reply-To: X-Scanned-By: MIMEDefang 2.79 on 10.5.11.13 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=maxime.coquelin@redhat.com X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit Subject: Re: [dpdk-stable] [PATCH 2/3] app/testpmd: fix RSS hash type update X-BeenThere: stable@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: patches for DPDK stable branches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: stable-bounces@dpdk.org Sender: "stable" On 9/16/21 9:41 AM, Li, Xiaoyun wrote: > Hi > >> -----Original Message----- >> From: Maxime Coquelin >> Sent: Thursday, September 16, 2021 15:34 >> To: Li, Xiaoyun ; Nélio Laranjeiro >> ; Yigit, Ferruh >> Cc: Andrew Rybchenko ; dev@dpdk.org; Xia, >> Chenbo ; amorenoz@redhat.com; >> david.marchand@redhat.com; michaelba@nvidia.com; viacheslavo@nvidia.com; >> stable@dpdk.org; Shahaf Shuler >> Subject: Re: [dpdk-stable] [PATCH 2/3] app/testpmd: fix RSS hash type update >> >> Hi Xiaoyun, >> >> On 9/16/21 5:03 AM, Li, Xiaoyun wrote: >>> Hi >>> >>>> -----Original Message----- >>>> From: stable On Behalf Of Nélio Laranjeiro >>>> Sent: Tuesday, September 14, 2021 15:20 >>>> To: Maxime Coquelin ; Yigit, Ferruh >>>> >>>> Cc: Andrew Rybchenko ; dev@dpdk.org; >>>> Xia, Chenbo ; amorenoz@redhat.com; >>>> david.marchand@redhat.com; michaelba@nvidia.com; >>>> viacheslavo@nvidia.com; stable@dpdk.org; Shahaf Shuler >>>> >>>> Subject: Re: [dpdk-stable] [PATCH 2/3] app/testpmd: fix RSS hash type >>>> update >>>> >>>> +Shahaf, >>>> >>>> Hi Maxime, >>>> >>>> On Mon, Sep 13, 2021 at 11:41:04AM +0200, Maxime Coquelin wrote: >>>>> Hi Nélio, >>>>> >>>>> On 9/10/21 4:16 PM, Nélio Laranjeiro wrote: >>>>>> On Fri, Sep 10, 2021 at 01:06:53PM +0300, Andrew Rybchenko wrote: >>>>>>> On 9/10/21 12:57 PM, Maxime Coquelin wrote: >>>>>>>> >>>>>>>> >>>>>>>> On 9/10/21 11:51 AM, Andrew Rybchenko wrote: >>>>>>>>> On 9/10/21 12:17 PM, Maxime Coquelin wrote: >>>>>>>>>> port_rss_hash_key_update() initializes rss_conf with the RSS >>>>>>>>>> hash type and key provided by the user, but it calls >>>>>>>>>> rte_eth_dev_rss_hash_conf_get() before calling >>>>>>>>>> rte_eth_dev_rss_hash_update(), which overides the parsed config >>>>>>>>>> with current NIC's config. >>>>>>>>>> >>>>>>>>>> While the RSS key value is set again after, this is not the >>>>>>>>>> case of the key length and the type of hash. >>>>>>>>>> >>>>>>>>>> There is no need to read the RSS config from the NIC, let's >>>>>>>>>> just try to set the user defined one. >>>>>>>>>> >>>>>>>>>> Fixes: 8205e241b2b0 ("app/testpmd: add missing type to RSS hash >>>>>>>>>> commands") >>>>>>>>>> Cc: stable@dpdk.org >>>>>>>>>> Cc: nelio.laranjeiro@6wind.com >>>>>>>>>> >>>>>>>>>> Signed-off-by: Maxime Coquelin >>>>>>>>>> --- >>>>>>>>>> app/test-pmd/config.c | 8 ++------ >>>>>>>>>> 1 file changed, 2 insertions(+), 6 deletions(-) >>>>>>>>>> >>>>>>>>>> diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c >>>>>>>>>> index >>>>>>>>>> 31d8ba1b91..451bda53b1 100644 >>>>>>>>>> --- a/app/test-pmd/config.c >>>>>>>>>> +++ b/app/test-pmd/config.c >>>>>>>>>> @@ -2853,18 +2853,14 @@ port_rss_hash_key_update(portid_t >>>> port_id, char rss_type[], uint8_t *hash_key, >>>>>>>>>> int diag; >>>>>>>>>> unsigned int i; >>>>>>>>>> >>>>>>>>>> - rss_conf.rss_key = NULL; >>>>>>>>>> + rss_conf.rss_key = hash_key; >>>>>>>>>> rss_conf.rss_key_len = hash_key_len; >>>>>>>>>> rss_conf.rss_hf = 0; >>>>>>>>>> for (i = 0; rss_type_table[i].str; i++) { >>>>>>>>>> if (!strcmp(rss_type_table[i].str, rss_type)) >>>>>>>>>> rss_conf.rss_hf = rss_type_table[i].rss_type; >>>>>>>>>> } >>>>>>>>>> - diag = rte_eth_dev_rss_hash_conf_get(port_id, &rss_conf); >>>>>>>>>> - if (diag == 0) { >>>>>>>>>> - rss_conf.rss_key = hash_key; >>>>>>>>>> - diag = rte_eth_dev_rss_hash_update(port_id, >>>> &rss_conf); >>>>>>>>>> - } >>>>>>>>>> + diag = rte_eth_dev_rss_hash_update(port_id, &rss_conf); >>>>>>>>> >>>>>>>>> I'm not 100% sure, but I'd say the intent above could be to >>>>>>>>> update key only as the function name says. I.e. keep rss_hf as >>>>>>>>> is. That could be the reason to get first. >>> >>> +1 >>> The intent is only updating rss key. That's why to get_rss_conf first. >>> At least for all intel devices. RSS key is a global value for all rss_hf. >>> >>> And since the intent is to only update the key value. I don't think this patch >> makes sense since you're changing rss_hf which will break current test cases. >>> And before 8205e241b2b01c, the command is what we want. >> port_rss_hash_key_update(portid_t port_id, uint8_t *hash_key) only updates >> key. >>> >>> But if mlx has the configuration of config key for each rss_type then the code >> should remain to the current code in which keylen and rss_hf will be overridden >> to the correct value intel wants and mlx has their own configuration for specific >> rss type. >>> But to be honest, I checked mlx5. I don't see they have this kind of >> configuration. Need to confirm with their driver maintainer though. >>> >>> But anyway, from my point of view, I prefer to revert what 8205e241b2b01c0 >> does and remove rss_hf, rss_key_len setting in this command if mlx doesn't have >> key update for specific rss type. Otherwise, remain what it's like wight now. >> >> For the main branch, we could revert 8205e241b2b01c0, but in my opinion, we >> need to keep the hash_key_length otherwise it could lead to out of bounds >> accesses if the key passed by the user is shorter than the key length in use by the >> driver. >> >> Note that it would also imply changes in the DTS and CIs tests that make use of >> this command. We would also need to introduce a new command to be able to >> set the rss hash types, and rename the existing one to be key- specific. >> Otherwise we have no way with testpmd to configure RSS properly. Given there >> does not seem to have any driver that leverages RSS HF/Key pairs, maybe the >> simple thing is to just do what I did in this patch. >> >> For stable, I don't think this is a good idea to change testpmd commands, as it >> could break users setups. > > I don’t think so. This command is used for key setting only like the name says. And users use this command only setting key. What you did in this patch actually will break test cases results. > > And change hash type you can use command "port config all rss eth|vlan|tcp|...". Ok, I did not find this command as I was trying completion with 'port config 0 rss'. It might be good to to change it so that we can configure it per port, but not a big deal. Remains the key length that should be passed, otherwise can have out of bounds accesses, no? And we still need to update DTS if we revert 8205e241b2b01c0. Thanks, Maxime > >> >> Thanks, >> Maxime >> >>> >>> BRs >>> Xiaoyun >>> >>>>>> >>>>>> True, >>>>>> >>>>>>>> I think that was the intial purpose of the command, but patch >>>>>>>> 8205e241b2b0 added setting the hash type as mandatory. There are >>>>>>>> no other command to configure the hash type from testpmd AFAICT. >>>>>> >>>>>> Also for the same initial purpose, some NIC have an hash key per >>>>>> protocol, by default it uses the same key for all of them but it >>>>>> can be configured individually making for example key0 for all >>>>>> protocols expect >>>>>> IPv4 which uses key1. >>>>> >>>>> Thanks for the info, I have looked at most drivers but didn't found >>>>> one that support this feature, could you give some pointer? >>>> >>>> Well, I have done the modification at that time for MLX5 PMD, since I >>>> left DPDK in 2018 I don't know if they still support such >>>> configuration from this API or if they fully moved to rte_flow. >>>> >>>>> Given how the drivers implément the callback, do you agree with the >>>>> fix, or do you have something else in mind? >>>> >>>> I cannot answer if this get() is mandatory, this predates my arrival >>>> on DPDK (original commit written in 2014), looking at DPDK state on >>>> commit >>>> f79959ea1504 ("app/testpmd: allow to configure RSS hash key"). >>>> Maybe someone from Intel can help, eventually you can contact PMD >>>> maintainers to review this patch. >>>> >>>> Regards, >>>> Nélio >>>> >>>>> Thanks, >>>>> Maxime >>>>> >>>>>>>> Also, even without 8205e241b2b0, the function was broken because >>>>>>>> the key length was overiden. >>>>>>> >>>>>>> I see, many thanks for explanations. >>>>>> >>>>> >>>> >>>> -- >>>> Nélio Laranjeiro >>>> 6WIND >>> >