From mboxrd@z Thu Jan 1 00:00:00 1970 From: Scott Feldman Subject: Re: [net-next-2.6 PATCH 1/2] Add ndo_set_vf_port_profile Date: Mon, 26 Apr 2010 17:03:23 -0700 Message-ID: References: Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Cc: "netdev@vger.kernel.org" , "chrisw@redhat.com" , "arnd@arndb.de" , "Williams, Mitch A" To: Scott Feldman , "Rose, Gregory V" , David Miller Return-path: Received: from sj-iport-1.cisco.com ([171.71.176.70]:33608 "EHLO sj-iport-1.cisco.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752046Ab0D0ADY (ORCPT ); Mon, 26 Apr 2010 20:03:24 -0400 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On 4/26/10 4:21 PM, "Scott Feldman" wrote: >> It seems to me that this: >> >> ip link show [ DEVICE ] >> >> should at least return the number of VFs so >> that you can make sure the subsequent usage of this: > > Yes, I believe that's there today: > > NLA_PUT_U32(skb, IFLA_NUM_VF, dev_num_vf(dev->dev.parent)); > > The number of VFs is returned in RTM_GETLINK. But, it's only returned if: > > if (dev->netdev_ops->ndo_get_vf_config && dev->dev.parent) > > For my proposal, I'll need to return IFLA_NUM_VF unconditionally so callers > can get num VFs. Hmmm...seems IFLA_NUM_VF assumes a PCI device supporting SR-IOV when it uses dev_num_vf(). I think a better option would have been to query the device for the number of VFs, without assuming SR-IOV or even PCI. I see a ndo_get_num_vf() coming... -scott