From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-ej1-f45.google.com (mail-ej1-f45.google.com [209.85.218.45]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 131502F83 for ; Fri, 23 Apr 2021 14:04:17 +0000 (UTC) Received: by mail-ej1-f45.google.com with SMTP id r9so74075010ejj.3 for ; Fri, 23 Apr 2021 07:04:17 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=tessares-net.20150623.gappssmtp.com; s=20150623; h=subject:to:cc:references:from:message-id:date:user-agent :mime-version:in-reply-to:content-language:content-transfer-encoding; bh=YOE8zwYy5j7sWC6DfZVirq0Juw5ukZCq6oGff2g4xGQ=; b=nfc0Q6j1c7NIc3gCyAtsymkt5aOq9SFnKCLBO3jeVpmT/JxydwaUPV2i5StaEq+KJi TLamHKKBPuXjGv99a14b09Dd9b0EQUxtzJI2j9hfugJ4OfYIMDyue/UVppVOGahYNkf7 UxL7PTFWUbiUJcpJKjPvFrbVTtB52w9q7p3gz/+79lFZmjb0cz1Ws4GObOeqNSJvWCLO IRN2miHmBSRfY1zHaxSfijFNPfbbehcnshat2JAuYnZbS40veMzZgsKTXFTt/2MHpiKu yJGvPjfPYBMdl9weP6VvYMVhCaiHrclJpmBYZAXxX43X9vgZqAG5VYmIU7QA4k8AzoKV ojrw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:subject:to:cc:references:from:message-id:date :user-agent:mime-version:in-reply-to:content-language :content-transfer-encoding; bh=YOE8zwYy5j7sWC6DfZVirq0Juw5ukZCq6oGff2g4xGQ=; b=ls1KVugbDYpdvxm5wzsA7+6IuAa0q+acfiygWDOON2YmggKAjBbMW3E8SUHTwlb5lR kBlix8Qzekqhll4LYH9MCQrwR60xkUnuncYILJGkw6xykEN6/QaDqZDAOKIQaZIuNf8y +7bgo0Q/RgwP5ld2OtQHE+2q+2hKZ0EPkGKHWgHiDyGoBrbV/SaU1p05O9fqfPD6uH1K WX3T3gYiEJSnGA1Qg+4eogjWBm1zXNR4DpVu91W7wbgx6znl4cGiohn58pHftDVhfXoy qM1R7+NDonnp2zINWu9CZMi2CzOHqTwS3VTGazaq+rwVbfV2oTxGvgF0Ok9lbCOnVBRW IlxA== X-Gm-Message-State: AOAM533at3+iAVKwJyKlmU7Okr9ODcrMlDKl7+/vxPq4e258KTOf2p4s lnKKVPfVBY8OS//cliX0O2KcMzPI4B/2/Q== X-Google-Smtp-Source: ABdhPJxwWcSGnPlxe3wTaOZPIK3Vx6FLPbYjrGds6IV210fdQbYACT5AbI6TrpwzmbqYjtzQi7UaoQ== X-Received: by 2002:a17:906:a155:: with SMTP id bu21mr4446479ejb.400.1619186656308; Fri, 23 Apr 2021 07:04:16 -0700 (PDT) Received: from tsr-lap-08.nix.tessares.net ([2a02:578:85b0:e00:d068:c16e:598a:1275]) by smtp.gmail.com with ESMTPSA id hg19sm4052534ejc.24.2021.04.23.07.04.15 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Fri, 23 Apr 2021 07:04:16 -0700 (PDT) Subject: Re: [PATCH] mptcp: create listening socket only if signal flag is set on To: Geliang Tang Cc: Jianguo Wu , mptcp@lists.linux.dev References: <1b50d699-45ec-e747-983a-1966f64f8e1c@163.com> <681e0810-fbbc-daaf-054a-0e83038b4d12@163.com> <3c3cf25d-aa4c-d522-102b-ebdd3b767cc0@tessares.net> From: Matthieu Baerts Message-ID: Date: Fri, 23 Apr 2021 16:04:15 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.8.1 X-Mailing-List: mptcp@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-GB Content-Transfer-Encoding: 8bit Hi Geliang, On 23/04/2021 10:59, Geliang Tang wrote: > Matthieu Baerts 于2021年4月23日周五 下午4:21写道： >> On 23/04/2021 09:41, Jianguo Wu wrote: >>> I think this is better, we can't assume that all user-space tools works correctly. >> >> I was thinking a bit about that. With the proposed modifications, it >> means that a server will have to send an ADD_ADDR if it wants to accept >> subflows on another port. I guess there will be use-cases where we will >> not want to send ADD_ADDR because we are in a controlled environment or >> we don't want to announce a public ADD_ADDR+port on a specific network. >> >> Do we have to support that? Maybe safer to restrict actions for now and >> allow that later if there are needs, no? >> >> In other words, I guess we should accept the proposed modifications. >> Any other points of view? :) > > I think the proposed modifications is not enough. > > If the port is set with non-signal flag (like subflow), we should not only > stop creating listening socket, but also stop appending this address. > > Since our code can't deal with the ADD_SUBFLOW+port case (like > "pm_nl_ctl add 10.0.0.1 flag subflow port 1000"). Local address's port will > re-set as the main socket's port number in our code. So set port to a local > address is useless, we should add the flag check to avoid this case. Indeed, that seems to make sense! Thanks for checking this! Cheers, Matt -- Tessares | Belgium | Hybrid Access Solutions www.tessares.net