| Message ID | c46e29b557e54652cce569e3348ed0385eac1469.1553170807.git.mkubecek@suse.cz |
|---|---|
| State | Changes Requested |
| Delegated to: | David Miller |
| Headers | show |
| Series | ethtool netlink interface, part 1 | expand |
On Thu, 21 Mar 2019 14:40:21 +0100 (CET) Michal Kubecek <mkubecek@suse.cz> wrote: > Permanent hardware address of a network device was traditionally provided > via ethtool ioctl interface but as Jiri Pirko pointed out in a review of > ethtool netlink interface, rtnetlink is much more suitable for it so let's > add it to the RTM_NEWLINK message. > > As permanent address is not modifiable, reject userspace requests > containing IFLA_PERM_ADDRESS attribute. > > Note: we already provide permanent hardware address for bond slaves; > unfortunately we cannot drop that attribute for backward compatibility > reasons. > > Signed-off-by: Michal Kubecek <mkubecek@suse.cz> > --- > include/uapi/linux/if_link.h | 1 + > net/core/rtnetlink.c | 4 ++++ > 2 files changed, 5 insertions(+) > > diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h > index 5b225ff63b48..351ef746b8b0 100644 > --- a/include/uapi/linux/if_link.h > +++ b/include/uapi/linux/if_link.h > @@ -167,6 +167,7 @@ enum { > IFLA_NEW_IFINDEX, > IFLA_MIN_MTU, > IFLA_MAX_MTU, > + IFLA_PERM_ADDRESS, > __IFLA_MAX > }; > > diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c > index a51cab95ba64..a72e8f4d777b 100644 > --- a/net/core/rtnetlink.c > +++ b/net/core/rtnetlink.c > @@ -1026,6 +1026,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev, > + nla_total_size(4) /* IFLA_CARRIER_DOWN_COUNT */ > + nla_total_size(4) /* IFLA_MIN_MTU */ > + nla_total_size(4) /* IFLA_MAX_MTU */ > + + nla_total_size(MAX_ADDR_LEN) /* IFLA_PERM_ADDRESS */ > + 0; > } > > @@ -1683,6 +1684,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, > nla_put_s32(skb, IFLA_NEW_IFINDEX, new_ifindex) < 0) > goto nla_put_failure; > > + if (nla_put(skb, IFLA_PERM_ADDRESS, dev->addr_len, dev->perm_addr)) > + goto nla_put_failure; > > rcu_read_lock(); > if (rtnl_fill_link_af(skb, dev, ext_filter_mask)) > @@ -1742,6 +1745,7 @@ static const struct nla_policy ifla_policy[IFLA_MAX+1] = { > [IFLA_CARRIER_DOWN_COUNT] = { .type = NLA_U32 }, > [IFLA_MIN_MTU] = { .type = NLA_U32 }, > [IFLA_MAX_MTU] = { .type = NLA_U32 }, > + [IFLA_PERM_ADDRESS] = { .type = NLA_REJECT }, > }; > > static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = { +1 this looks good, please also send iproute2 patch to display it.
On Thu, 21 Mar 2019 14:40:21 +0100 (CET), Michal Kubecek wrote: > Permanent hardware address of a network device was traditionally provided > via ethtool ioctl interface but as Jiri Pirko pointed out in a review of > ethtool netlink interface, rtnetlink is much more suitable for it so let's > add it to the RTM_NEWLINK message. > > As permanent address is not modifiable, reject userspace requests > containing IFLA_PERM_ADDRESS attribute. > > Note: we already provide permanent hardware address for bond slaves; > unfortunately we cannot drop that attribute for backward compatibility > reasons. > > Signed-off-by: Michal Kubecek <mkubecek@suse.cz> > --- > include/uapi/linux/if_link.h | 1 + > net/core/rtnetlink.c | 4 ++++ > 2 files changed, 5 insertions(+) > > diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h > index 5b225ff63b48..351ef746b8b0 100644 > --- a/include/uapi/linux/if_link.h > +++ b/include/uapi/linux/if_link.h > @@ -167,6 +167,7 @@ enum { > IFLA_NEW_IFINDEX, > IFLA_MIN_MTU, > IFLA_MAX_MTU, > + IFLA_PERM_ADDRESS, > __IFLA_MAX > }; > > diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c > index a51cab95ba64..a72e8f4d777b 100644 > --- a/net/core/rtnetlink.c > +++ b/net/core/rtnetlink.c > @@ -1026,6 +1026,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev, > + nla_total_size(4) /* IFLA_CARRIER_DOWN_COUNT */ > + nla_total_size(4) /* IFLA_MIN_MTU */ > + nla_total_size(4) /* IFLA_MAX_MTU */ > + + nla_total_size(MAX_ADDR_LEN) /* IFLA_PERM_ADDRESS */ > + 0; > } > > @@ -1683,6 +1684,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, > nla_put_s32(skb, IFLA_NEW_IFINDEX, new_ifindex) < 0) > goto nla_put_failure; > > + if (nla_put(skb, IFLA_PERM_ADDRESS, dev->addr_len, dev->perm_addr)) Should we check the perm_addr is non-zero, i.e. it has been filled in by the driver? > + goto nla_put_failure; > > rcu_read_lock(); > if (rtnl_fill_link_af(skb, dev, ext_filter_mask)) > @@ -1742,6 +1745,7 @@ static const struct nla_policy ifla_policy[IFLA_MAX+1] = { > [IFLA_CARRIER_DOWN_COUNT] = { .type = NLA_U32 }, > [IFLA_MIN_MTU] = { .type = NLA_U32 }, > [IFLA_MAX_MTU] = { .type = NLA_U32 }, > + [IFLA_PERM_ADDRESS] = { .type = NLA_REJECT }, > }; > > static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = {
From: Michal Kubecek <mkubecek@suse.cz> Date: Thu, 21 Mar 2019 14:40:21 +0100 (CET) > Permanent hardware address of a network device was traditionally provided > via ethtool ioctl interface but as Jiri Pirko pointed out in a review of > ethtool netlink interface, rtnetlink is much more suitable for it so let's > add it to the RTM_NEWLINK message. > > As permanent address is not modifiable, reject userspace requests > containing IFLA_PERM_ADDRESS attribute. > > Note: we already provide permanent hardware address for bond slaves; > unfortunately we cannot drop that attribute for backward compatibility > reasons. > > Signed-off-by: Michal Kubecek <mkubecek@suse.cz> Jakub asked if we should check the perm_addr and if it is set. I would say no, in order to be consistent with what ethtool does which is that it does not check.
On Thu, Mar 21, 2019 at 01:35:05PM -0700, Jakub Kicinski wrote: > On Thu, 21 Mar 2019 14:40:21 +0100 (CET), Michal Kubecek wrote: > > Permanent hardware address of a network device was traditionally provided > > via ethtool ioctl interface but as Jiri Pirko pointed out in a review of > > ethtool netlink interface, rtnetlink is much more suitable for it so let's > > add it to the RTM_NEWLINK message. > > > > As permanent address is not modifiable, reject userspace requests > > containing IFLA_PERM_ADDRESS attribute. > > > > Note: we already provide permanent hardware address for bond slaves; > > unfortunately we cannot drop that attribute for backward compatibility > > reasons. > > > > Signed-off-by: Michal Kubecek <mkubecek@suse.cz> > > --- > > include/uapi/linux/if_link.h | 1 + > > net/core/rtnetlink.c | 4 ++++ > > 2 files changed, 5 insertions(+) > > > > diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h > > index 5b225ff63b48..351ef746b8b0 100644 > > --- a/include/uapi/linux/if_link.h > > +++ b/include/uapi/linux/if_link.h > > @@ -167,6 +167,7 @@ enum { > > IFLA_NEW_IFINDEX, > > IFLA_MIN_MTU, > > IFLA_MAX_MTU, > > + IFLA_PERM_ADDRESS, > > __IFLA_MAX > > }; > > > > diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c > > index a51cab95ba64..a72e8f4d777b 100644 > > --- a/net/core/rtnetlink.c > > +++ b/net/core/rtnetlink.c > > @@ -1026,6 +1026,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev, > > + nla_total_size(4) /* IFLA_CARRIER_DOWN_COUNT */ > > + nla_total_size(4) /* IFLA_MIN_MTU */ > > + nla_total_size(4) /* IFLA_MAX_MTU */ > > + + nla_total_size(MAX_ADDR_LEN) /* IFLA_PERM_ADDRESS */ > > + 0; > > } > > > > @@ -1683,6 +1684,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, > > nla_put_s32(skb, IFLA_NEW_IFINDEX, new_ifindex) < 0) > > goto nla_put_failure; > > > > + if (nla_put(skb, IFLA_PERM_ADDRESS, dev->addr_len, dev->perm_addr)) > > Should we check the perm_addr is non-zero, i.e. it has been filled in > by the driver? It makes sense logically but there is IMHO one practical problem: if we omit the attribute when the address is zero (i.e. not set or maybe a device which does not have permanent address), userspace won't be able to distinguish this case from an older kernel not implementing the attribute. For the iproute2 patch, I plan to show the permanent address in the output of "ip link show" only if it differs from current address. I think it would be desirable to distinguish "current address is the permanent one" from "no/unknown permanent address" but absence of the attribute could also mean "kernel does not provide the information". Michal
diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index 5b225ff63b48..351ef746b8b0 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -167,6 +167,7 @@ enum { IFLA_NEW_IFINDEX, IFLA_MIN_MTU, IFLA_MAX_MTU, + IFLA_PERM_ADDRESS, __IFLA_MAX }; diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index a51cab95ba64..a72e8f4d777b 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1026,6 +1026,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev, + nla_total_size(4) /* IFLA_CARRIER_DOWN_COUNT */ + nla_total_size(4) /* IFLA_MIN_MTU */ + nla_total_size(4) /* IFLA_MAX_MTU */ + + nla_total_size(MAX_ADDR_LEN) /* IFLA_PERM_ADDRESS */ + 0; } @@ -1683,6 +1684,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, nla_put_s32(skb, IFLA_NEW_IFINDEX, new_ifindex) < 0) goto nla_put_failure; + if (nla_put(skb, IFLA_PERM_ADDRESS, dev->addr_len, dev->perm_addr)) + goto nla_put_failure; rcu_read_lock(); if (rtnl_fill_link_af(skb, dev, ext_filter_mask)) @@ -1742,6 +1745,7 @@ static const struct nla_policy ifla_policy[IFLA_MAX+1] = { [IFLA_CARRIER_DOWN_COUNT] = { .type = NLA_U32 }, [IFLA_MIN_MTU] = { .type = NLA_U32 }, [IFLA_MAX_MTU] = { .type = NLA_U32 }, + [IFLA_PERM_ADDRESS] = { .type = NLA_REJECT }, }; static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = {
Permanent hardware address of a network device was traditionally provided via ethtool ioctl interface but as Jiri Pirko pointed out in a review of ethtool netlink interface, rtnetlink is much more suitable for it so let's add it to the RTM_NEWLINK message. As permanent address is not modifiable, reject userspace requests containing IFLA_PERM_ADDRESS attribute. Note: we already provide permanent hardware address for bond slaves; unfortunately we cannot drop that attribute for backward compatibility reasons. Signed-off-by: Michal Kubecek <mkubecek@suse.cz> --- include/uapi/linux/if_link.h | 1 + net/core/rtnetlink.c | 4 ++++ 2 files changed, 5 insertions(+)