From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-13.7 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_CR_TRAILER,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED, USER_AGENT_SANE_1 autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id D9C9CC63777 for ; Thu, 3 Dec 2020 12:25:08 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 703EF22244 for ; Thu, 3 Dec 2020 12:25:08 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1730402AbgLCMZH (ORCPT ); Thu, 3 Dec 2020 07:25:07 -0500 Received: from hqnvemgate25.nvidia.com ([216.228.121.64]:13712 "EHLO hqnvemgate25.nvidia.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728469AbgLCMZG (ORCPT ); Thu, 3 Dec 2020 07:25:06 -0500 Received: from hqmail.nvidia.com (Not Verified[216.228.121.13]) by hqnvemgate25.nvidia.com (using TLS: TLSv1.2, AES256-SHA) id ; Thu, 03 Dec 2020 04:24:26 -0800 Received: from mtl-vdi-166.wap.labs.mlnx (10.124.1.5) by HQMAIL107.nvidia.com (172.20.187.13) with Microsoft SMTP Server (TLS) id 15.0.1473.3; Thu, 3 Dec 2020 12:24:24 +0000 Date: Thu, 3 Dec 2020 14:24:21 +0200 From: Eli Cohen To: "Michael S. Tsirkin" CC: Jason Wang , Cindy Lu , , Subject: Re: [PATCH] vdpa/mlx5: Use random MAC for the vdpa net instance Message-ID: <20201203122421.GB38007@mtl-vdi-166.wap.labs.mlnx> References:

<20201202055714.GA224423@mtl-vdi-166.wap.labs.mlnx> <20201202041518-mutt-send-email-mst@kernel.org> <3e32ef6d-83c9-5866-30e5-f6eeacd5044d@redhat.com> <20201202165932-mutt-send-email-mst@kernel.org> <20201203064928.GA27404@mtl-vdi-166.wap.labs.mlnx> <20201203054330-mutt-send-email-mst@kernel.org> <20201203120929.GA38007@mtl-vdi-166.wap.labs.mlnx> <20201203071414-mutt-send-email-mst@kernel.org> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Disposition: inline Content-Transfer-Encoding: quoted-printable In-Reply-To: <20201203071414-mutt-send-email-mst@kernel.org> User-Agent: Mutt/1.9.5 (bf161cf53efb) (2018-04-13) X-Originating-IP: [10.124.1.5] X-ClientProxiedBy: HQMAIL105.nvidia.com (172.20.187.12) To HQMAIL107.nvidia.com (172.20.187.13) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=nvidia.com; s=n1; t=1606998266; bh=NwcmaiKNlMitO7tSakJ9HNs7b6P4137S7PCAUu+AePc=; h=Date:From:To:CC:Subject:Message-ID:References:MIME-Version: Content-Type:Content-Disposition:Content-Transfer-Encoding: In-Reply-To:User-Agent:X-Originating-IP:X-ClientProxiedBy; b=aizC9/zXcC8oYC9P31+BOSlHKeLBJzxW/QxET3E6HAhqaP1oBYNz0M/4CezvRlf7F 0SI8SNduqrjrnPNVL6S0vkDE8a/nL6xHNPh/VPHOXmArWoDEqwXLvr+90ON1fvAOJU J8hSuOP9XYWuFNIqKlRkQrQ3EsZDged8W6bDZ2++iCbeoJpKFjgqZ7DA5dgkyjh9XH TtwV0LfJbMiBUwdGY18TQtSMQ47kE2G3ldQHvnbspR/WOwggTH6tcjfz7X2dICQpIo qswqqQwtDKGn0+d1YUe3RSdHB6P6ElmLpEscFv+pNwFG6q9nhgW91zrF/wx7ZoL+aU 6Q+rwAWN/5txw== Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Dec 03, 2020 at 07:15:37AM -0500, Michael S. Tsirkin wrote: > On Thu, Dec 03, 2020 at 02:09:29PM +0200, Eli Cohen wrote: > > On Thu, Dec 03, 2020 at 05:44:17AM -0500, Michael S. Tsirkin wrote: > > > On Thu, Dec 03, 2020 at 08:49:28AM +0200, Eli Cohen wrote: > > > > On Wed, Dec 02, 2020 at 05:00:22PM -0500, Michael S. Tsirkin wrote: > > > > > On Wed, Dec 02, 2020 at 09:48:25PM +0800, Jason Wang wrote: > > > > > >=20 > > > > > > On 2020/12/2 =E4=B8=8B=E5=8D=885:23, Michael S. Tsirkin wrote: > > > > > > > On Wed, Dec 02, 2020 at 07:57:14AM +0200, Eli Cohen wrote: > > > > > > > > On Wed, Dec 02, 2020 at 12:18:36PM +0800, Jason Wang wrote: > > > > > > > > > On 2020/12/1 =E4=B8=8B=E5=8D=885:23, Cindy Lu wrote: > > > > > > > > > > On Mon, Nov 30, 2020 at 11:33 PM Michael S. Tsirkin wrote: > > > > > > > > > > > On Mon, Nov 30, 2020 at 06:41:45PM +0800, Cindy Lu wr= ote: > > > > > > > > > > > > On Mon, Nov 30, 2020 at 5:33 PM Michael S. Tsirkin<= mst@redhat.com> wrote: > > > > > > > > > > > > > On Mon, Nov 30, 2020 at 11:27:59AM +0200, Eli Coh= en wrote: > > > > > > > > > > > > > > On Mon, Nov 30, 2020 at 04:00:51AM -0500, Micha= el S. Tsirkin wrote: > > > > > > > > > > > > > > > On Mon, Nov 30, 2020 at 08:27:46AM +0200, Eli= Cohen wrote: > > > > > > > > > > > > > > > > On Sun, Nov 29, 2020 at 03:08:22PM -0500, M= ichael S. Tsirkin wrote: > > > > > > > > > > > > > > > > > On Sun, Nov 29, 2020 at 08:43:51AM +0200,= Eli Cohen wrote: > > > > > > > > > > > > > > > > > > We should not try to use the VF MAC add= ress as that is used by the > > > > > > > > > > > > > > > > > > regular (e.g. mlx5_core) NIC implementa= tion. Instead, use a random > > > > > > > > > > > > > > > > > > generated MAC address. > > > > > > > > > > > > > > > > > >=20 > > > > > > > > > > > > > > > > > > Suggested by: Cindy Lu > > > > > > > > > > > > > > > > > > Fixes: 1a86b377aa21 ("vdpa/mlx5: Add VD= PA driver for supported mlx5 devices") > > > > > > > > > > > > > > > > > > Signed-off-by: Eli Cohen > > > > > > > > > > > > > > > > > I didn't realise it's possible to use VF = in two ways > > > > > > > > > > > > > > > > > with and without vdpa. > > > > > > > > > > > > > > > > Using a VF you can create quite a few resou= rces, e.g. send queues > > > > > > > > > > > > > > > > recieve queues, virtio_net queues etc. So y= ou can possibly create > > > > > > > > > > > > > > > > several instances of vdpa net devices and n= ic net devices. > > > > > > > > > > > > > > > >=20 > > > > > > > > > > > > > > > > > Could you include a bit more description = on the failure > > > > > > > > > > > > > > > > > mode? > > > > > > > > > > > > > > > > Well, using the MAC address of the nic vpor= t is wrong since that is the > > > > > > > > > > > > > > > > MAC of the regular NIC implementation of ml= x5_core. > > > > > > > > > > > > > > > Right but ATM it doesn't coexist with vdpa so= what's the problem? > > > > > > > > > > > > > > >=20 > > > > > > > > > > > > > > This call is wrong: mlx5_query_nic_vport_mac_a= ddress() > > > > > > > > > > > > > >=20 > > > > > > > > > > > > > > > > > Is switching to a random mac for such an = unusual > > > > > > > > > > > > > > > > > configuration really justified? > > > > > > > > > > > > > > > > Since I can't use the NIC's MAC address, I = have two options: > > > > > > > > > > > > > > > > 1. To get the MAC address as was chosen by = the user administering the > > > > > > > > > > > > > > > > NIC. This should invoke the set_config= callback. Unfortunately this > > > > > > > > > > > > > > > > is not implemented yet. > > > > > > > > > > > > > > > >=20 > > > > > > > > > > > > > > > > 2. Use a random MAC address. This is OK sin= ce if (1) is implemented it > > > > > > > > > > > > > > > > can always override this random config= uration. > > > > > > > > > > > > > > > >=20 > > > > > > > > > > > > > > > > > It looks like changing a MAC could break = some guests, > > > > > > > > > > > > > > > > > can it not? > > > > > > > > > > > > > > > > >=20 > > > > > > > > > > > > > > > > No, it will not. The current version of mlx= 5 VDPA does not allow regular > > > > > > > > > > > > > > > > NIC driver and VDPA to co-exist. I have pat= ches ready that enable that > > > > > > > > > > > > > > > > from steering point of view. I will post th= em here once other patches on > > > > > > > > > > > > > > > > which they depend will be merged. > > > > > > > > > > > > > > > >=20 > > > > > > > > > > > > > > > > https://patchwork.ozlabs.org/project/netdev= /patch/20201120230339.651609-12-saeedm@nvidia.com/ > > > > > > > > > > > > > > > Could you be more explicit on the following p= oints: > > > > > > > > > > > > > > > - which configuration is broken ATM (as in, t= wo device have identical > > > > > > > > > > > > > > > macs? any other issues)? > > > > > > > > > > > > > > The only wrong thing is the call to mlx5_query= _nic_vport_mac_address(). > > > > > > > > > > > > > > It's not breaking anything yet is wrong. The ra= ndom MAC address setting > > > > > > > > > > > > > > is required for the steering patches. > > > > > > > > > > > > > Okay so I'm not sure the Fixes tag at least is ap= propriate if it's a > > > > > > > > > > > > > dependency of a new feature. > > > > > > > > > > > > >=20 > > > > > > > > > > > > > > > - why won't device MAC change from guest poin= t of view? > > > > > > > > > > > > > > >=20 > > > > > > > > > > > > > > It's lack of implementation in qemu as far as I= know. > > > > > > > > > > > > > Sorry not sure I understand. What's not implement= ed in QEMU? > > > > > > > > > > > > >=20 > > > > > > > > > > > > HI Michael, there are some bug in qemu to set_confi= g, this will fix in future, > > > > > > > > > > > > But this patch is still needed, because without thi= s patch the mlx > > > > > > > > > > > > driver will give an 0 mac address to qemu > > > > > > > > > > > > and qemu will overwrite the default mac address. T= his will cause traffic down. > > > > > > > > > > > Hmm the patch description says VF mac address, not 0 = address. Confused. > > > > > > > > > > > If there's no mac we can clear VIRTIO_NET_F_MAC and h= ave guest > > > > > > > > > > > use a random value ... > > > > > > > > > I'm not sure this can work for all types of vDPA (e.g it = could not be a > > > > > > > > > learning bridge in the swtich). > > > > > > > > >=20 > > > > > > > > >=20 > > > > > > > > > > hi Michael=EF=BC=8C > > > > > > > > > > I have tried as your suggestion, seems even remove the > > > > > > > > > > VIRTIO_NET_F_MAC the qemu will still call get_cinfig an= d overwrite the > > > > > > > > > > default address in VM, > > > > > > > > > This looks a bug in qemu, in guest driver we had: > > > > > > > > >=20 > > > > > > > > > =C2=A0=C2=A0=C2=A0 /* Configuration may specify what MAC= to use.=C2=A0 Otherwise random. */ > > > > > > > > > =C2=A0=C2=A0=C2=A0 if (virtio_has_feature(vdev, VIRTIO_N= ET_F_MAC)) > > > > > > > > > =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 virtio_cread_bytes= (vdev, > > > > > > > > > =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0 offsetof(struct virtio_net_config, mac), > > > > > > > > > =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0 dev->dev_addr, dev->addr_len); > > > > > > > > > =C2=A0=C2=A0=C2=A0 else > > > > > > > > > =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 eth_hw_addr_random= (dev); > > > > > > > > >=20 > > > > > > > > >=20 > > > > > > > > > > this process is like > > > > > > > > > > vdpa _init -->qemu call get_config ->mlx driver will gi= ve an mac > > > > > > > > > > address with all 0--> > > > > > > > > > > qemu will not check this mac address and use it --> ove= rwrite the mac > > > > > > > > > > address in qemu > > > > > > > > > >=20 > > > > > > > > > > So for my understanding there are several method to fix= this problem > > > > > > > > > >=20 > > > > > > > > > > 1, qemu check the mac address, if the mac address is al= l 0, qemu will > > > > > > > > > > ignore it and set the random mac address to mlx driver. > > > > > > > > > So my understanding is that, if mac address is all 0, vDP= A parent should not > > > > > > > > > advertise VIRTIO_NET_F_MAC. And qemu should emulate this = feature as you did: > > > > > > > > Thinking it over, at least in mlx5, I should always adverti= se > > > > > > > > VIRTIO_NET_F_MAC and set a non zero MAC value. The source o= f the MAC can > > > > > > > > be either randomly generated value by mlx5_vdpa or by a man= agement tool. > > > > > > > > This is important becauase we should not let the VM modify = the MAC. If > > > > > > > > we do it can set a MAC value identical to the mlx5 NIC driv= er and can > > > > > > > > kidnap traffic that was not destined to it. > > > > > > > >=20 > > > > > > > > In addition, when VIRTIO_NET_F_MAC is published, attempts t= o change the > > > > > > > > MAC address from the VM should result in error. > > > > > > > That is not what the spec says though. > > > > > > > VIRTIO_NET_F_MAC only says whether mac is valid in the config= space. > > > > > > > Whether guest can control that depends on VIRTIO_NET_F_CTRL_M= AC_ADDR: > > > > > > >=20 > > > > > > > The VIRTIO_NET_CTRL_MAC_ADDR_SET command is used to set the = default MAC address which rx > > > > > > > filtering accepts (and if VIRTIO_NET_F_MAC_ADDR has been neg= otiated, this will be reflected in mac in > > > > > > > config space). > > > > > > > The command-specific-data for VIRTIO_NET_CTRL_MAC_ADDR_SET i= s the 6-byte MAC address. > > > > > >=20 > > > > > >=20 > > > > > > Consider VIRTIO_NET_CTRL_MAC_ADDR_SET is not supported now. Wha= t Eli > > > > > > proposed here should work? > > > > > >=20 > > > > > > Thanks > > > > > >=20 > > > > >=20 > > > > > We can have management set a MAC address. Randomizing it in kerne= l > > > > > does not seem like a reasonable policy to me ... > > > > >=20 > > > >=20 > > > > This manangement should be the VDPA tool that Parav is pushing. We = can > > > > use it to set a MAC chosen by the user. The mlx5 vdpa driver can th= en > > > > use that MAC instead of randomizing a value. If no admin value is g= iven > > > > we can use a random MAC. > > >=20 > > > IIUC in this model devices are created by this tool, right? > > > Why not require the MAC when device is created? > > >=20 > >=20 > > It is mentioned in Parav's patchset that this will be coming in a > > subsequent patch to his vdpa tool.=20 >=20 > So I think kernel has two options: > - require a mac when device is created, we supply it to guest Yes, the driver should always set VIRTIO_NET_F_MAC and provide a MAC - either random or whatever configured using the vdpa too. > - allow guest to set a mac >=20 This cannot work today. Moreover, if's a problem to allow the guest to set the MAC becuase it can take over the regular mlx5 NIC's address. > it's ok to support both ... >=20 >=20 >=20 > > > > >=20 > > > > > > >=20 > > > > > > >=20 > > > > > > >=20 > > > > > > >=20 > > > > >=20 > > >=20 >=20