From: Jason Wang <jasowang@redhat.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: davem@davemloft.net, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [net-next RFC 6/8] macvtap: allow TUNSETIFF to create multiqueue device
Date: Fri, 24 May 2013 13:36:50 +0800 [thread overview]
Message-ID: <519EFC72.1010405@redhat.com> (raw)
In-Reply-To: <20130523114556.GC17993@redhat.com>
On 05/23/2013 07:45 PM, Michael S. Tsirkin wrote:
> On Thu, May 23, 2013 at 11:12:31AM +0800, Jason Wang wrote:
>> Though the queue were in fact created by open(), we still need to add this check
>> to be compatible with tuntap which can let mgmt software use a single API to
>> manage queues. This patch only validates the device name and moves the TUNSETIFF
>> to a helper.
>>
>> Signed-off-by: Jason Wang <jasowang@redhat.com>
>> ---
>> drivers/net/macvtap.c | 50 ++++++++++++++++++++++++++++++++++++++----------
>> 1 files changed, 39 insertions(+), 11 deletions(-)
>>
>> diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
>> index 5fd341c..e3f9344 100644
>> --- a/drivers/net/macvtap.c
>> +++ b/drivers/net/macvtap.c
>> @@ -867,6 +867,7 @@ out:
>> return ret;
>> }
>>
>> +
> Please don't add empty lines like this :).
Sure.
>
>> static struct macvlan_dev *macvtap_get_vlan(struct macvtap_queue *q)
>> {
>> struct macvlan_dev *vlan;
>> @@ -884,6 +885,43 @@ static void macvtap_put_vlan(struct macvlan_dev *vlan)
>> dev_put(vlan->dev);
>> }
>>
>> +static int macvtap_set_iff(struct file *file, struct ifreq __user *ifr_u)
>> +{
>> + struct macvtap_queue *q = file->private_data;
>> + struct net *net = current->nsproxy->net_ns;
>> + struct inode *inode = file_inode(file);
>> + struct net_device *dev, *dev2;
>> + struct ifreq ifr;
>> +
>> + if (copy_from_user(&ifr, ifr_u, sizeof(struct ifreq)))
>> + return -EFAULT;
>> +
>> + if (ifr.ifr_flags & IFF_MULTI_QUEUE) {
> So why do we validate the name?
> Pls add a comment.
Ok. The validation is to prevent the userspace create the queues on the
wrong device. Not sure whether or not this is needed for single queue case.
>
>> + dev = __dev_get_by_name(net, ifr.ifr_name);
>> + if (!dev)
>> + return -EINVAL;
>> +
>> + dev2 = dev_get_by_macvtap_minor(iminor(inode));
>> + if (!dev2)
>> + return -EINVAL;
>> +
>> + if (dev != dev2) {
>> + dev_put(dev2);
>> + return -EINVAL;
>> + }
>> +
>> + dev_put(dev2);
>> + }
>> +
>> + if ((ifr.ifr_flags & ~(IFF_VNET_HDR | IFF_MULTI_QUEUE)) !=
>> + (IFF_NO_PI | IFF_TAP))
>> + return -EINVAL;
>> + else
>> + q->flags = ifr.ifr_flags;
>> +
>> + return 0;
>> +}
>> +
>> /*
>> * provide compatibility with generic tun/tap interface
>> */
>> @@ -902,17 +940,7 @@ static long macvtap_ioctl(struct file *file, unsigned int cmd,
>>
>> switch (cmd) {
>> case TUNSETIFF:
>> - /* ignore the name, just look at flags */
>> - if (get_user(u, &ifr->ifr_flags))
>> - return -EFAULT;
>> -
>> - ret = 0;
>> - if ((u & ~IFF_VNET_HDR) != (IFF_NO_PI | IFF_TAP))
>> - ret = -EINVAL;
>> - else
>> - q->flags = u;
>> -
>> - return ret;
>> + return macvtap_set_iff(file, ifr);
>>
>> case TUNGETIFF:
>> vlan = macvtap_get_vlan(q);
>> --
>> 1.7.1
next prev parent reply other threads:[~2013-05-24 5:36 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-05-23 3:12 [net-next RFC 0/8] multiqueue API support for macvtap Jason Wang
2013-05-23 3:12 ` [net-next RFC 1/8] macvlan: switch to use IS_ENABLED() Jason Wang
2013-05-23 11:31 ` Michael S. Tsirkin
2013-05-23 3:12 ` [net-next RFC 2/8] macvtap: return -EBADFD when TUNGETIFF fails Jason Wang
2013-05-23 11:54 ` Michael S. Tsirkin
2013-05-24 6:28 ` Jason Wang
2013-05-23 3:12 ` [net-next RFC 3/8] macvtap: introduce macvtap_get_vlan() Jason Wang
2013-05-23 15:11 ` Sergei Shtylyov
2013-05-24 6:21 ` Jason Wang
2013-05-23 3:12 ` [net-next RFC 4/8] macvlan: reduce the max number of taps to 8 Jason Wang
2013-05-23 6:37 ` Michael S. Tsirkin
2013-05-24 5:14 ` Jason Wang
2013-05-23 3:12 ` [net-next RFC 5/8] macvtap: eliminate linear search Jason Wang
2013-05-23 11:41 ` Michael S. Tsirkin
2013-05-24 5:33 ` Jason Wang
2013-05-23 3:12 ` [net-next RFC 6/8] macvtap: allow TUNSETIFF to create multiqueue device Jason Wang
2013-05-23 11:45 ` Michael S. Tsirkin
2013-05-24 5:36 ` Jason Wang [this message]
2013-05-23 3:12 ` [net-next RFC 7/8] macvtap: add TUNSETQUEUE ioctl Jason Wang
2013-05-23 11:52 ` Michael S. Tsirkin
2013-05-24 6:19 ` Jason Wang
2013-05-23 3:12 ` [net-next RFC 8/8] macvtap: enable multiqueue flag Jason Wang
2013-05-23 11:53 ` [net-next RFC 0/8] multiqueue API support for macvtap Michael S. Tsirkin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=519EFC72.1010405@redhat.com \
--to=jasowang@redhat.com \
--cc=davem@davemloft.net \
--cc=linux-kernel@vger.kernel.org \
--cc=mst@redhat.com \
--cc=netdev@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).