Fri, Jul 19, 2019 at 06:31:35PM CEST, step...@networkplumber.org wrote: >On Fri, 19 Jul 2019 13:00:22 +0200 >Jiri Pirko <j...@resnulli.us> wrote: > >> From: Jiri Pirko <j...@mellanox.com> >> >> In the past, there was repeatedly discussed the IFNAMSIZ (16) limit for >> netdevice name length. Now when we have PF and VF representors >> with port names like "pfXvfY", it became quite common to hit this limit: >> 0123456789012345 >> enp131s0f1npf0vf6 >> enp131s0f1npf0vf22 >> >> Udev cannot rename these interfaces out-of-the-box and user needs to >> create custom rules to handle them. >> >> Also, udev has multiple schemes of netdev names. From udev code: >> * Type of names: >> * b<number> - BCMA bus core number >> * c<bus_id> - bus id of a grouped CCW or CCW >> device, >> * with all leading zeros stripped >> [s390] >> * o<index>[n<phys_port_name>|d<dev_port>] >> * - on-board device index number >> * s<slot>[f<function>][n<phys_port_name>|d<dev_port>] >> * - hotplug slot index number >> * x<MAC> - MAC address >> * [P<domain>]p<bus>s<slot>[f<function>][n<phys_port_name>|d<dev_port>] >> * - PCI geographical location >> * >> [P<domain>]p<bus>s<slot>[f<function>][u<port>][..][c<config>][i<interface>] >> * - USB port number chain >> * v<slot> - VIO slot number (IBM PowerVM) >> * a<vendor><model>i<instance> - Platform bus ACPI instance id >> * i<addr>n<phys_port_name> - Netdevsim bus address and port >> name >> >> One device can be often renamed by multiple patterns at the >> same time (e.g. pci address/mac). >> >> This patchset introduces alternative names for network interfaces. >> Main goal is to: >> 1) Overcome the IFNAMSIZ limitation >> 2) Allow to have multiple names at the same time (multiple udev patterns) >> 3) Allow to use alternative names as handle for commands >> >> See following examples. >> >> $ ip link >> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode >> DEFAULT group default qlen 1000 >> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 >> 3: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT >> group default qlen 1000 >> link/ether 7e:a2:d4:b8:91:7a brd ff:ff:ff:ff:ff:ff >> >> -> Add alternative names for dummy0: >> >> $ ip link altname add dummy0 name someothername >> $ ip link altname add dummy0 name someotherveryveryveryverylongname >> $ ip link >> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode >> DEFAULT group default qlen 1000 >> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 >> 3: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT >> group default qlen 1000 >> link/ether 7e:a2:d4:b8:91:7a brd ff:ff:ff:ff:ff:ff >> altname someothername >> altname someotherveryveryveryverylongname >> >> -> Add bridge brx, add it's alternative name and use alternative names to >> do enslavement. >> >> $ ip link add name brx type bridge >> $ ip link altname add brx name mypersonalsuperspecialbridge >> $ ip link set someotherveryveryveryverylongname master >> mypersonalsuperspecialbridge >> $ ip link >> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode >> DEFAULT group default qlen 1000 >> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 >> 3: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop master brx state DOWN mode >> DEFAULT group default qlen 1000 >> link/ether 7e:a2:d4:b8:91:7a brd ff:ff:ff:ff:ff:ff >> altname someothername >> altname someotherveryveryveryverylongname >> 4: brx: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT >> group default qlen 1000 >> link/ether 7e:a2:d4:b8:91:7a brd ff:ff:ff:ff:ff:ff >> altname mypersonalsuperspecialbridge >> >> -> Add ipv4 address to the bridge using alternative name: >> >> $ ip addr add 192.168.0.1/24 dev mypersonalsuperspecialbridge >> $ ip addr show mypersonalsuperspecialbridge >> 4: brx: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default >> qlen 1000 >> link/ether 7e:a2:d4:b8:91:7a brd ff:ff:ff:ff:ff:ff >> altname mypersonalsuperspecialbridge >> inet 192.168.0.1/24 scope global brx >> valid_lft forever preferred_lft forever >> >> -> Delete one of dummy0 alternative names: >> >> $ ip link altname del someotherveryveryveryverylongname >> $ ip link >> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode >> DEFAULT group default qlen 1000 >> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 >> 3: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop master brx state DOWN mode >> DEFAULT group default qlen 1000 >> link/ether 7e:a2:d4:b8:91:7a brd ff:ff:ff:ff:ff:ff >> altname someothername >> 4: brx: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT >> group default qlen 1000 >> link/ether 7e:a2:d4:b8:91:7a brd ff:ff:ff:ff:ff:ff >> altname mypersonalsuperspecialbridge >> >> TODO: >> - notifications for alternative names add/removal >> - sanitization of add/del cmds (similar to get link) >> - test more usecases and write selftests >> - extend support for other netlink ifaces (ovs for example) >> - add sysfs symlink altname->basename? >> >> Jiri Pirko (7): >> net: procfs: use index hashlist instead of name hashlist >> net: introduce name_node struct to be used in hashlist >> net: rtnetlink: add commands to add and delete alternative ifnames >> net: rtnetlink: put alternative names to getlink message >> net: rtnetlink: unify the code in __rtnl_newlink get dev with the rest >> net: rtnetlink: introduce helper to get net_device instance by ifname >> net: rtnetlink: add possibility to use alternative names as message >> handle >> >> include/linux/netdevice.h | 14 ++- >> include/uapi/linux/if.h | 1 + >> include/uapi/linux/if_link.h | 3 + >> include/uapi/linux/rtnetlink.h | 7 ++ >> net/core/dev.c | 152 ++++++++++++++++++++++---- >> net/core/net-procfs.c | 4 +- >> net/core/rtnetlink.c | 192 +++++++++++++++++++++++++++++---- >> security/selinux/nlmsgtab.c | 4 +- >> 8 files changed, 334 insertions(+), 43 deletions(-) >> > >You might want to add altname/ object property in sysfs? >I.e > /sys/class/net/eth0/altname/
Sysfs representation is one point on my TODO list. I didn't look much into this, but yeah, that would work. I also want to have symlinks altname->basename. >
