From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from out4-smtp.messagingengine.com (out4-smtp.messagingengine.com [66.111.4.28]) by dpdk.org (Postfix) with ESMTP id AC19F1B01D for ; Wed, 24 Jan 2018 09:19:51 +0100 (CET) Received: from compute1.internal (compute1.nyi.internal [10.202.2.41]) by mailout.nyi.internal (Postfix) with ESMTP id 455BF22718; Wed, 24 Jan 2018 03:19:51 -0500 (EST) Received: from frontend1 ([10.202.2.160]) by compute1.internal (MEProxy); Wed, 24 Jan 2018 03:19:51 -0500 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=monjalon.net; h= cc:content-transfer-encoding:content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to:x-me-sender :x-me-sender:x-sasl-enc; s=mesmtp; bh=oDzKq5jHM/Ftu8qUBZLBbfNc/6 V49oRpwa/uwBNDSrk=; b=g+bjg684xwjLPHWY/KZbAWN+KW8PMDP0hY7KfNH8fw yHJGEVeOGKaXLH2IaMvR+YkxONlyVOsFycCfTz4TA6AiBpTK5vP5LzIXS2DYwAEc XFrGswvhlk2+fU+dHIILr20hcow8pch3uM58BsW9LY0g/OpYitZkUt8PMBvM9FkR E= DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d= messagingengine.com; h=cc:content-transfer-encoding:content-type :date:from:in-reply-to:message-id:mime-version:references :subject:to:x-me-sender:x-me-sender:x-sasl-enc; s=fm1; bh=oDzKq5 jHM/Ftu8qUBZLBbfNc/6V49oRpwa/uwBNDSrk=; b=TECCeTydZGVXkMM26OE1dD wpEVZgZjwQ17ac7Ygnb+ws4tdxwvCNIcWKeT5S2JDT5CMFEuM7wrIkpPAzXLUYBu MqjSDo0XqqoUODxsLtXbwRN4c/vT9sjfccJHoIGTxMNcpqdphugXNk/0gbAf5nsu 0QYWiXWBDG/A79RAGU+byCA8kPrsk2Vm7Ar/d06TI85OSjXQUxEGTy0o2YQp9onW 3GqgC/phKQUEIfHioRqKC5vNlVOyuh+CWOr5MXuljdbwYLjFpo417j9zBZB5H1xK UKyW2z7Be5KJe3WT2tglhO5RBY2gqaqL2AzoRaKdl0Y8CHj8NFtNiC9+Kff8lW/Q == X-ME-Sender: Received: from xps.localnet (184.203.134.77.rev.sfr.net [77.134.203.184]) by mail.messagingengine.com (Postfix) with ESMTPA id EA0147E354; Wed, 24 Jan 2018 03:19:50 -0500 (EST) From: Thomas Monjalon To: Yuanhan Liu Cc: =?ISO-8859-1?Q?Ga=EBtan?= Rivet , Ferruh Yigit , dev@dpdk.org Date: Wed, 24 Jan 2018 09:19:10 +0100 Message-ID: <1680850.DZ6KxSLkSi@xps> In-Reply-To: <20180124064339.GT29540@yliu-mob> References: <1516114218-21501-1-git-send-email-yliu@fridaylinux.org> <17059484.9s6tGoQGou@xps> <20180124064339.GT29540@yliu-mob> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="iso-8859-1" Subject: Re: [dpdk-dev] [PATCH] doc: document the new devargs syntax 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: , X-List-Received-Date: Wed, 24 Jan 2018 08:19:51 -0000 24/01/2018 07:43, Yuanhan Liu: > On Tue, Jan 23, 2018 at 03:29:34PM +0100, Thomas Monjalon wrote: > > 23/01/2018 13:46, Yuanhan Liu: > > > On Thu, Jan 18, 2018 at 10:46:23AM +0100, Ga=EBtan Rivet wrote: > > > > On Thu, Jan 18, 2018 at 09:46:29AM +0100, Thomas Monjalon wrote: > > > > > 18/01/2018 08:35, Yuanhan Liu: > > > > > > On Wed, Jan 17, 2018 at 12:34:08PM +0000, Ferruh Yigit wrote: > > > > > > > So does it make sense to separate them logically? Perhaps as = "device identifier" > > > > > > > and "device args". > > > > > >=20 > > > > > > Then I think it returns back to the old issue: how could we ide= ntify a > > > > > > port when the bus id (say BDF for PCI bus) is not enough for id= entifying > > > > > > a port? Such case could happen when a single NIC has 2 ports sh= aring > > > > > > the same BDF. It could also happen with the VF representors tha= t will > > > > > > be introduced shortly. > > > > >=20 > > > > > Yes, the device matching syntax must include bus category, class = category > > > > > and driver category. So any device can be identified in future. > > > > >=20 > > > > > But I think Ferruh is talking about separating device matching > > > > > (which is described in this proposal) and device settings > > > > > (which are usually mixed in -w and --vdev options). > > > > > I agree there are different things and may be separate. > > > > > They could share the same syntax (bus/class/driver) but be separa= te > > > > > with a semicolon: > > > > > matching;settings > > > > > > > > Can you give an example? > > >=20 > > > Let's take port addition in OVS-DPDK as an example. It happens in 2 > > > steps: > > > - port lookup (if port is already probed) > > > - dev attachment (if lookup fails) > > >=20 > > > And also let's assume we need probe a ConnectX-3 port. Note that for > > > ConnectX-3, there are 2 ports sharing the same PCI addr. Thus, PCI > > > BDF is not enough. And let's assume we use another extra property > > > "port". > > >=20 > > > If the proposal described in this patch is being used, the devarg > > > would look like following: > > >=20 > > > bus=3Dpci,id=3D04:00.0/class=3Deth,port=3D0/driver=3Dmlx4,mlx4_ar= g_A=3Dval,... > > >=20 > > > Then "bus=3Dpci,id=3D04:00.0/class=3Deth,port=3D0" will be used for l= ookup, > > > It means we are looking for a port with PCI BDF =3D=3D 04:00.0 AND > > > port =3D=3D 0 (the first port of the 2 ports). > > >=20 > > > Note that in my proposal the driver category is not intended for look= up. > > > If any properties needed be looked in the driver category, they would > > > probably need be elevated to the class category. > >=20 > > It is not my thought. > > I think we should be able to use bus, class and driver properties for l= ookup. > > We can imagine doing a lookup on a driver specific id, which is not > > candidate to elevation to the class category. > >=20 > > > If port not found, then the whole string will be used for dev attachm= ent. > > > It means we are attaching a port with PCI BDF =3D=3D 04.00.0 AND > > > port =3D=3D 0 (the 2nd port will not be attached). > > >=20 > > >=20 > > > And here is how the devargs would look like if "matching;settings" is > > > being used: > > >=20 > > > bus=3Dpci,id=3D04:00.0/class=3Deth,port=3D0;bus=3Dpci,id=3D04:00.= 0/class=3Deth,port=3D0/driver=3Dmlx4,mlx4_arg_A=3Dval,... > > >=20 > > > The part before ";" will be used for lookup and the later part will be > > > used for attachment. It should work. It just looks redundant. > >=20 > > It does not have to be redundant. > > It can be: > > bus=3Dpci,id=3D04:00.0/class=3Deth,port=3D0;driver=3Dmlx4,mlx4_arg1=3D= settings1,... >=20 > I knew you would make such reply :) > Then there is a contradiction. According your suggestion, the "port=3D0" = belongs > to the matching section, but it also has to be used in the settings secti= on. If port=3D0 is matched, it is already set, right? Why it needs to be in settings? > > Another example, setting the MAC address: > > bus=3Dpci,id=3D04:00.0/class=3Deth,port=3D0;class=3Deth,mac=3D00:11:22= :33:44:55 >=20 > What's the scenario it will be used? And who is going to parse it? It can be used on command line for whitelisting a device and let the user change its MAC address. The matching is parsed by the PCI driver + ethdev, here. As mac is a property of ethdev (class=3Deth), this part must be parsed by e= thdev.