Merging upstream version 102.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

5 files changed, 1080 insertions, 0 deletions

diff --git a/modules.d/45net-lib/dhcp-root.sh b/modules.d/45net-lib/dhcp-root.sh
new file mode 100755
index 0000000..3f11221
--- /dev/null
+++ b/modules.d/45net-lib/dhcp-root.sh
@@ -0,0 +1,30 @@
+#!/bin/sh
+
+# This script is sourced, so root should be set. But let's be paranoid
+[ -z "$root" ] && root=$(getarg root=)
+
+if [ -z "$netroot" ]; then
+    for netroot in $(getargs netroot=); do
+        [ "$netroot" = "dhcp" ] && break
+        [ "$netroot" = "dhcp6" ] && break
+    done
+    [ "$netroot" = "dhcp" ] || [ "$netroot" = "dhcp6" ] || unset netroot
+fi
+
+if [ "$root" = "dhcp" ] || [ "$root" = "dhcp6" ] || [ "$netroot" = "dhcp" ] || [ "$netroot" = "dhcp6" ]; then
+    # Tell ip= checker that we need dhcp
+    # shellcheck disable=SC2034
+    NEEDDHCP="1"
+
+    # Done, all good!
+    # shellcheck disable=SC2034
+    rootok=1
+    if [ "$netroot" != "dhcp" ] && [ "$netroot" != "dhcp6" ]; then
+        netroot=$root
+    fi
+
+    # Shut up init error check
+    [ -z "$root" ] && root="dhcp"
+    # shellcheck disable=SC2016
+    echo '[ -d $NEWROOT/proc -o -e /dev/root ]' > "$hookdir"/initqueue/finished/dhcp.sh
+fi
diff --git a/modules.d/45net-lib/ifname-genrules.sh b/modules.d/45net-lib/ifname-genrules.sh
new file mode 100755
index 0000000..b9b95c4
--- /dev/null
+++ b/modules.d/45net-lib/ifname-genrules.sh
@@ -0,0 +1,32 @@
+#!/bin/sh
+
+# if there are no ifname parameters, just use NAME=KERNEL
+if ! getarg ifname= > /dev/null; then
+    return
+fi
+
+command -v parse_ifname_opts > /dev/null || . /lib/net-lib.sh
+
+{
+    for p in $(getargs ifname=); do
+        parse_ifname_opts "$p"
+
+        if [ -f /tmp/ifname-"$ifname_mac" ]; then
+            read -r oldif < /tmp/ifname-"$ifname_mac"
+        fi
+        if [ -f /tmp/ifname-"$ifname_if" ]; then
+            read -r oldmac < /tmp/ifname-"$ifname_if"
+        fi
+        if [ -n "$oldif" -a -n "$oldmac" -a "$oldif" = "$ifname_if" -a "$oldmac" = "$ifname_mac" ]; then
+            # skip same ifname= declaration
+            continue
+        fi
+
+        [ -n "$oldif" ] && warn "Multiple interface names specified for MAC $ifname_mac: $oldif"
+        [ -n "$oldmac" ] && warn "Multiple MAC specified for $ifname_if: $oldmac"
+
+        printf 'SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="%s", ATTR{type}=="1", NAME="%s"\n' "$ifname_mac" "$ifname_if"
+        echo "$ifname_if" > /tmp/ifname-"$ifname_mac"
+        echo "$ifname_mac" > /tmp/ifname-"$ifname_if"
+    done
+} >> /etc/udev/rules.d/80-ifname.rules
diff --git a/modules.d/45net-lib/module-setup.sh b/modules.d/45net-lib/module-setup.sh
new file mode 100755
index 0000000..8053aa6
--- /dev/null
+++ b/modules.d/45net-lib/module-setup.sh
@@ -0,0 +1,20 @@
+#!/bin/bash
+
+check() {
+    return 255
+}
+
+depends() {
+    echo base
+    return 0
+}
+
+install() {
+    inst_script "$moddir/netroot.sh" "/sbin/netroot"
+    inst_simple "$moddir/net-lib.sh" "/lib/net-lib.sh"
+    inst_hook pre-udev 50 "$moddir/ifname-genrules.sh"
+    inst_hook cmdline 91 "$moddir/dhcp-root.sh"
+    inst_multiple ip sed awk grep pgrep tr
+    inst_multiple -o arping arping2
+    dracut_need_initqueue
+}
diff --git a/modules.d/45net-lib/net-lib.sh b/modules.d/45net-lib/net-lib.sh
new file mode 100755
index 0000000..9d88e0d
--- /dev/null
+++ b/modules.d/45net-lib/net-lib.sh
@@ -0,0 +1,906 @@
+#!/bin/sh
+
+# shellcheck disable=SC2034
+IFNETFILE="/tmp/bootnetif"
+
+is_ip() {
+    echo "$1" | {
+        IFS=. read -r a b c d
+        test "$a" -ge 0 -a "$a" -le 255 \
+            -a "$b" -ge 0 -a "$b" -le 255 \
+            -a "$c" -ge 0 -a "$c" -le 255 \
+            -a "$d" -ge 0 -a "$d" -le 255 \
+            2> /dev/null
+    } && return 0
+    return 1
+}
+
+get_ip() {
+    local iface="$1" ip=""
+    ip=$(ip -f inet addr show "$iface")
+    ip=${ip%%/*}
+    ip=${ip##* }
+    echo "$ip"
+}
+
+iface_for_remote_addr() {
+    # shellcheck disable=SC2046
+    set -- $(ip route get to "$@" | sed 's/.*\bdev\b//p;q')
+    echo "$1"
+}
+
+iface_for_ip() {
+    # shellcheck disable=SC2046
+    set -- $(ip addr show to "$@")
+    echo "${2%:}"
+}
+
+iface_for_mac() {
+    local interface=""
+    local mac
+    mac="$(echo "$@" | sed 'y/ABCDEF/abcdef/')"
+    for interface in /sys/class/net/*; do
+        if [ "$(cat "$interface"/address)" = "$mac" ]; then
+            echo "${interface##*/}"
+        fi
+    done
+}
+
+# get the iface name for the given identifier - either a MAC, IP, or iface name
+iface_name() {
+    case "$1" in
+        ??:??:??:??:??:?? | ??-??-??-??-??-??) iface_for_mac "$1" ;;
+        *:*:* | *.*.*.*) iface_for_ip "$1" ;;
+        *) echo "$1" ;;
+    esac
+}
+
+# list the configured interfaces
+configured_ifaces() {
+    local IFACES="" iface_id="" rv=1
+    [ -e "/tmp/net.ifaces" ] && read -r IFACES < /tmp/net.ifaces
+    if { pidof udevd || pidof systemd-udevd; } > /dev/null; then
+        for iface_id in $IFACES; do
+            printf "%s\n" "$(iface_name "$iface_id")"
+            rv=0
+        done
+    else
+        warn "configured_ifaces called before udev is running"
+        echo "$IFACES"
+        [ -n "$IFACES" ] && rv=0
+    fi
+    return $rv
+}
+
+all_ifaces_up() {
+    local iface="" IFACES=""
+    [ -e "/tmp/net.ifaces" ] && read -r IFACES < /tmp/net.ifaces
+    for iface in $IFACES; do
+        [ -e /tmp/net."$iface".up ] || return 1
+    done
+}
+
+all_ifaces_setup() {
+    local iface="" IFACES=""
+    [ -e "/tmp/net.ifaces" ] && read -r IFACES < /tmp/net.ifaces
+    for iface in $IFACES; do
+        [ -e /tmp/net."$iface".did-setup ] || return 1
+    done
+}
+
+get_netroot_ip() {
+    local prefix="" server="" rest=""
+    splitsep ":" "$1" prefix server rest
+    case $server in
+        [0-9]*\.[0-9]*\.[0-9]*\.[0-9]*)
+            echo "$server"
+            return 0
+            ;;
+    esac
+    return 1
+}
+
+ip_is_local() {
+    li=$(ip route get "$@" 2> /dev/null)
+    if [ -n "$li" ]; then
+        strstr "$li" " via " || return 0
+    fi
+    return 1
+}
+
+ifdown() {
+    local netif="$1"
+    # ip down/flush ensures that routing info goes away as well
+    ip link set "$netif" down
+    ip addr flush dev "$netif"
+    echo "#empty" > /etc/resolv.conf
+    rm -f -- /tmp/net."$netif".did-setup
+    [ -z "$DO_VLAN" ] \
+        && [ -e /sys/class/net/"$netif"/address ] \
+        && rm -f -- "/tmp/net.$(cat /sys/class/net/"$netif"/address).did-setup"
+    # TODO: send "offline" uevent?
+}
+
+setup_net() {
+    local netif="$1" f="" gw_ip="" netroot_ip="" iface="" IFACES=""
+    local _p
+    [ -e /tmp/net."$netif".did-setup ] && return
+    [ -z "$DO_VLAN" ] \
+        && [ -e /sys/class/net/"$netif"/address ] \
+        && [ -e "/tmp/net.$(cat /sys/class/net/"$netif"/address).did-setup" ] && return
+    [ -e "/tmp/net.ifaces" ] && read -r IFACES < /tmp/net.ifaces
+    [ -z "$IFACES" ] && IFACES="$netif"
+    # run the scripts written by ifup
+    # shellcheck disable=SC1090
+    [ -e /tmp/net."$netif".hostname ] && . /tmp/net."$netif".hostname
+    # shellcheck disable=SC1090
+    [ -e /tmp/net."$netif".override ] && . /tmp/net."$netif".override
+    # shellcheck disable=SC1090
+    [ -e /tmp/dhclient."$netif".dhcpopts ] && . /tmp/dhclient."$netif".dhcpopts
+    # set up resolv.conf
+    [ -e /tmp/net."$netif".resolv.conf ] \
+        && awk '!array[$0]++' /tmp/net."$netif".resolv.conf > /etc/resolv.conf
+    # shellcheck disable=SC1090
+    [ -e /tmp/net."$netif".gw ] && . /tmp/net."$netif".gw
+
+    # add static route
+    for _p in $(getargs rd.route); do
+        route_to_var "$_p" || continue
+        [ -n "$route_dev" ] && [ "$route_dev" != "$netif" ] && continue
+        ip route add "$route_mask" ${route_gw:+via $route_gw} ${route_dev:+dev $route_dev}
+        if strstr "$route_mask" ":"; then
+            printf -- "%s\n" "$route_mask ${route_gw:+via $route_gw} ${route_dev:+dev $route_dev}" \
+                > /tmp/net.route6."$netif"
+        else
+            printf -- "%s\n" "$route_mask ${route_gw:+via $route_gw} ${route_dev:+dev $route_dev}" \
+                > /tmp/net.route."$netif"
+        fi
+    done
+
+    # If a static route was necessary to reach the gateway, the
+    # first gateway setup call will have failed with
+    #     RTNETLINK answers: Network is unreachable
+    # Replace the default route again after static routes to cover
+    # this scenario.
+    # shellcheck disable=SC1090
+    [ -e /tmp/net."$netif".gw ] && . /tmp/net."$netif".gw
+
+    # Handle STP Timeout: arping the default gateway.
+    # (or the root server, if a) it's local or b) there's no gateway.)
+    # Note: This assumes that if no router is present the
+    # root server is on the same subnet.
+
+    # Get DHCP-provided router IP, or the cmdline-provided "gw=" argument
+    [ -n "$new_routers" ] && gw_ip=${new_routers%%,*}
+    [ -n "$gw" ] && gw_ip=$gw
+
+    # Get the "netroot" IP (if there's an IP address in there)
+    netroot_ip=$(get_netroot_ip "$netroot")
+
+    # try netroot if it's local (or there's no gateway)
+    if ip_is_local "$netroot_ip" || [ -z "$gw_ip" ]; then
+        dest="$netroot_ip"
+    else
+        dest="$gw_ip"
+    fi
+
+    unset layer2
+    if [ -f /sys/class/net/"$netif"/device/layer2 ]; then
+        read -r layer2 < /sys/class/net/"$netif"/device/layer2
+    fi
+
+    if [ "$layer2" != "0" ] && [ -n "$dest" ] && ! strstr "$dest" ":"; then
+        if command -v arping2 > /dev/null; then
+            arping2 -q -C 1 -c 60 -I "$netif" "$dest" || info "Resolving $dest via ARP on $netif failed"
+        else
+            arping -q -f -w 60 -I "$netif" "$dest" || info "Resolving $dest via ARP on $netif failed"
+        fi
+    fi
+    unset layer2
+
+    : > /tmp/net."$netif".did-setup
+    [ -z "$DO_VLAN" ] \
+        && [ -e /sys/class/net/"$netif"/address ] \
+        && : > "/tmp/net.$(cat /sys/class/net/"$netif"/address).did-setup"
+}
+
+save_netinfo() {
+    local netif="$1" IFACES="" f="" i=""
+    [ -e /tmp/net.ifaces ] && read -r IFACES < /tmp/net.ifaces
+    # Add $netif to the front of IFACES (if it's not there already).
+    set -- "$netif"
+    for i in $IFACES; do [ "$i" != "$netif" ] && set -- "$@" "$i"; done
+    IFACES="$*"
+    for i in $IFACES; do
+        for f in "/tmp/dhclient.$i."*; do
+            [ -f "$f" ] && cp -f "$f" /tmp/net."${f#/tmp/dhclient.}"
+        done
+    done
+    echo "$IFACES" > /tmp/.net.ifaces.new
+    mv /tmp/.net.ifaces.new /tmp/net.ifaces
+}
+
+set_ifname() {
+    local name="$1" mac="$2" num=-1 n=""
+    # if it's already set, return the existing name
+    for n in $(getargs ifname=); do
+        strstr "$n" "$mac" && echo "${n%%:*}" && return
+    done
+    [ ! -f "/tmp/set_ifname_$name" ] || read -r num < "/tmp/set_ifname_$name"
+    # otherwise, pick a new name and use that
+    while :; do
+        num=$((num + 1))
+        [ -e /sys/class/net/"$name"$num ] && continue
+        for n in $(getargs ifname=); do
+            [ "$name$num" = "${n%%:*}" ] && continue 2
+        done
+        break
+    done
+    echo "ifname=$name$num:$mac" >> /etc/cmdline.d/45-ifname.conf
+    echo "$num" > "/tmp/set_ifname_$name"
+    echo "$name$num"
+}
+
+# pxelinux provides macaddr '-' separated, but we need ':'
+fix_bootif() {
+    local macaddr="${1}"
+    local IFS='-'
+    # shellcheck disable=SC2086
+    macaddr=$(printf '%s:' ${macaddr})
+    macaddr=${macaddr%:}
+    # strip hardware type field from pxelinux
+    [ -n "${macaddr%??:??:??:??:??:??}" ] && macaddr=${macaddr#??:}
+    # return macaddr with lowercase alpha characters expected by udev
+    echo "$macaddr" | sed 'y/ABCDEF/abcdef/'
+}
+
+ibft_to_cmdline() {
+    local iface=""
+    modprobe -q iscsi_ibft
+    (
+        for iface in /sys/firmware/ibft/ethernet*; do
+            local mac="" dev=""
+            local dhcp="" ip="" gw="" mask="" hostname=""
+            local dns1 dns2
+
+            [ -e "${iface}"/mac ] || continue
+            read -r mac < "${iface}"/mac
+            [ -z "$mac" ] && continue
+            dev=$(set_ifname ibft "$mac")
+
+            [ -e /tmp/net."${dev}".has_ibft_config ] && continue
+
+            [ -e "${iface}"/flags ] && read -r flags < "${iface}"/flags
+            # Skip invalid interfaces
+            awk -- 'BEGIN { exit (!and('"$flags"',1)) }' || continue
+            # Skip interfaces not used for booting unless using multipath
+            if ! getargbool 0 rd.iscsi.mp; then
+                awk -- 'BEGIN { exit (!and('"$flags"',2)) }' || continue
+            fi
+            [ -e "${iface}"/dhcp ] && read -r dhcp < "${iface}"/dhcp
+            [ -e "${iface}"/origin ] && read -r origin < "${iface}"/origin
+            [ -e "${iface}"/ip-addr ] && read -r ip < "${iface}"/ip-addr
+
+            if [ -n "$ip" ]; then
+                case "$ip" in
+                    *.*.*.*)
+                        family=ipv4
+                        ;;
+                    *:*)
+                        family=ipv6
+                        ;;
+                esac
+            fi
+            if [ -n "$dhcp" ] || [ "$origin" -eq 3 ]; then
+                if [ "$family" = "ipv6" ]; then
+                    echo "ip=$dev:dhcp6"
+                else
+                    echo "ip=$dev:dhcp"
+                fi
+            elif [ -e "${iface}"/ip-addr ]; then
+                # skip not assigned ip addresses
+                [ "$ip" = "0.0.0.0" ] && continue
+                [ -e "${iface}"/gateway ] && read -r gw < "${iface}"/gateway
+                [ "$gw" = "0.0.0.0" ] && unset gw
+                [ -e "${iface}"/subnet-mask ] && read -r mask < "${iface}"/subnet-mask
+                [ -e "${iface}"/prefix-len ] && read -r prefix < "${iface}"/prefix-len
+                [ -e "${iface}"/primary-dns ] && read -r dns1 < "${iface}"/primary-dns
+                [ "$dns1" = "0.0.0.0" ] && unset dns1
+                [ -e "${iface}"/secondary-dns ] && read -r dns2 < "${iface}"/secondary-dns
+                [ "$dns2" = "0.0.0.0" ] && unset dns
+                [ -e "${iface}"/hostname ] && read -r hostname < "${iface}"/hostname
+                if [ "$family" = "ipv6" ]; then
+                    if [ -n "$ip" ]; then
+                        [ -n "$prefix" ] || prefix=128
+                        ip="[${ip}]"
+                        mask=$prefix
+                    fi
+                    if [ -n "$gw" ]; then
+                        gw="[${gw}]"
+                    fi
+                fi
+                if [ -n "$ip" ] && [ -n "$mask" -o -n "$prefix" ]; then
+                    echo "ip=$ip::$gw:$mask:$hostname:$dev:none${dns1:+:$dns1}${dns2:+:$dns2}"
+                else
+                    warn "${iface} does not contain a valid iBFT configuration"
+                    warn "ip-addr=$ip"
+                    warn "gateway=$gw"
+                    warn "subnet-mask=$mask"
+                    warn "hostname=$hostname"
+                fi
+            else
+                info "${iface} does not contain a valid iBFT configuration"
+                # shellcheck disable=SC2012
+                ls -l "${iface}" | vinfo
+            fi
+
+            if [ -e "${iface}"/vlan ]; then
+                read -r vlan < "${iface}"/vlan
+                if [ "$vlan" -ne "0" ]; then
+                    case "$vlan" in
+                        [0-9]*)
+                            echo "vlan=$dev.$vlan:$dev"
+                            echo "$mac" > /tmp/net."${dev}"."${vlan}".has_ibft_config
+                            ;;
+                        *)
+                            echo "vlan=$vlan:$dev"
+                            echo "$mac" > /tmp/net."${vlan}".has_ibft_config
+                            ;;
+                    esac
+                else
+                    echo "$mac" > /tmp/net."${dev}".has_ibft_config
+                fi
+            else
+                echo "$mac" > /tmp/net."${dev}".has_ibft_config
+            fi
+
+        done
+    ) >> /etc/cmdline.d/40-ibft.conf
+}
+
+parse_iscsi_root() {
+    local v
+    v=${1#iscsi:}
+
+    # extract authentication info
+    case "$v" in
+        *@*:*:*:*:*)
+            authinfo=${v%%@*}
+            v=${v#*@}
+            # allow empty authinfo to allow having an @ in iscsi_target_name like this:
+            # netroot=iscsi:@192.168.1.100::3260::iqn.2009-01.com.example:testdi@sk
+            if [ -n "$authinfo" ]; then
+                OLDIFS="$IFS"
+                IFS=:
+                # shellcheck disable=SC2086
+                set $authinfo
+                IFS="$OLDIFS"
+                if [ $# -gt 4 ]; then
+                    warn "Wrong authentication info in iscsi: parameter!"
+                    return 1
+                fi
+                iscsi_username=$1
+                iscsi_password=$2
+                if [ $# -gt 2 ]; then
+                    iscsi_in_username=$3
+                    iscsi_in_password=$4
+                fi
+            fi
+            ;;
+    esac
+
+    # extract target ip
+    case "$v" in
+        [[]*[]]:*)
+            iscsi_target_ip=${v#[[]}
+            iscsi_target_ip=${iscsi_target_ip%%[]]*}
+            v=${v#[[]"$iscsi_target_ip"[]]:}
+            ;;
+        *)
+            iscsi_target_ip=${v%%[:]*}
+            v=${v#"$iscsi_target_ip":}
+            ;;
+    esac
+
+    unset iscsi_target_name
+    # extract target name
+    case "$v" in
+        *:iqn.*)
+            iscsi_target_name=iqn.${v##*:iqn.}
+            v=${v%:iqn.*}:
+            ;;
+        *:eui.*)
+            iscsi_target_name=eui.${v##*:eui.}
+            v=${v%:eui.*}:
+            ;;
+        *:naa.*)
+            iscsi_target_name=naa.${v##*:naa.}
+            v=${v%:naa.*}:
+            ;;
+    esac
+
+    # parse the rest
+    OLDIFS="$IFS"
+    IFS=:
+    # shellcheck disable=SC2086
+    set $v
+    IFS="$OLDIFS"
+
+    iscsi_protocol=$1
+    shift # ignored
+    iscsi_target_port=$1
+    shift
+
+    if [ -n "$iscsi_target_name" ]; then
+        if [ $# -eq 3 ]; then
+            iscsi_iface_name=$1
+            shift
+        fi
+        if [ $# -eq 2 ]; then
+            iscsi_netdev_name=$1
+            shift
+        fi
+        iscsi_lun=$1
+        shift
+        if [ $# -ne 0 ]; then
+            warn "Invalid parameter in iscsi: parameter!"
+            return 1
+        fi
+        return 0
+    fi
+
+    if [ $# -gt 3 ] && [ -n "$1$2" ]; then
+        if [ -z "$3" ] || [ "$3" -ge 0 ] 2> /dev/null; then
+            iscsi_iface_name=$1
+            shift
+            iscsi_netdev_name=$1
+            shift
+        fi
+    fi
+
+    iscsi_lun=$1
+    shift
+
+    iscsi_target_name=$(printf "%s:" "$@")
+    iscsi_target_name=${iscsi_target_name%:}
+}
+
+ip_to_var() {
+    local v="${1}":
+    local i
+    set --
+    while [ -n "$v" ]; do
+        if [ "${v#\[*:*:*\]:}" != "$v" ]; then
+            # handle IPv6 address
+            i="${v%%\]:*}"
+            i="${i##\[}"
+            set -- "$@" "$i"
+            v=${v#\["$i"\]:}
+        else
+            set -- "$@" "${v%%:*}"
+            v=${v#*:}
+        fi
+    done
+
+    unset ip srv gw mask hostname dev autoconf macaddr mtu dns1 dns2
+
+    if [ $# -eq 0 ]; then
+        autoconf="error"
+        return 0
+    fi
+
+    if [ $# -eq 1 ]; then
+        # format: ip={dhcp|on|any|dhcp6|auto6|either6|single-dhcp}
+        # or
+        #         ip=<ipv4-address> means anaconda-style static config argument cluster
+        autoconf="$1"
+
+        if strglob "$autoconf" "*.*.*.*"; then
+            # ip=<ipv4-address> means anaconda-style static config argument cluster:
+            # ip=<ip> gateway=<gw> netmask=<nm> hostname=<host> mtu=<mtu>
+            # ksdevice={link|bootif|ibft|<MAC>|<ifname>}
+            ip="$autoconf"
+            gw=$(getarg gateway=)
+            mask=$(getarg netmask=)
+            hostname=$(getarg hostname=)
+            dev=$(getarg ksdevice=)
+            autoconf="none"
+            mtu=$(getarg mtu=)
+
+            # handle special values for ksdevice
+            case "$dev" in
+                bootif | BOOTIF) dev=$(fix_bootif "$(getarg BOOTIF=)") ;;
+                link) dev="" ;; # FIXME: do something useful with this
+                ibft) dev="" ;; # ignore - ibft is handled elsewhere
+            esac
+        fi
+        return 0
+    fi
+
+    if [ "$2" = "dhcp" -o "$2" = "on" -o "$2" = "any" -o "$2" = "dhcp6" -o "$2" = "auto6" -o "$2" = "either6" ]; then
+        # format: ip=<interface>:{dhcp|on|any|dhcp6|auto6}[:[<mtu>][:<macaddr>]]
+        [ -n "$1" ] && dev="$1"
+        [ -n "$2" ] && autoconf="$2"
+        [ -n "$3" ] && mtu=$3
+        if [ -z "$5" ]; then
+            macaddr="$4"
+        else
+            macaddr="${4}:${5}:${6}:${7}:${8}:${9}"
+        fi
+        return 0
+    fi
+
+    # format: ip=<client-IP>:[<peer>]:<gateway-IP>:<netmask>:<client_hostname>:<interface>:{none|off|dhcp|on|any|dhcp6|auto6|ibft}:[:[<mtu>][:<macaddr>]]
+
+    [ -n "$1" ] && ip=$1
+    [ -n "$2" ] && srv=$2
+    [ -n "$3" ] && gw=$3
+    [ -n "$4" ] && mask=$4
+    [ -n "$5" ] && hostname=$5
+    [ -n "$6" ] && dev=$6
+    [ -n "$7" ] && autoconf=$7
+    case "$8" in
+        [0-9a-fA-F]*:* | [0-9]*.[0-9]*.[0-9]*.[0-9]*)
+            dns1="$8"
+            [ -n "$9" ] && dns2="$9"
+            ;;
+        [0-9]*)
+            mtu="$8"
+            if [ -n "${9}" -a -z "${10}" ]; then
+                macaddr="${9}"
+            elif [ -n "${9}" -a -n "${10}" -a -n "${11}" -a -n "${12}" -a -n "${13}" -a -n "${14}" ]; then
+                macaddr="${9}:${10}:${11}:${12}:${13}:${14}"
+            fi
+            ;;
+        *)
+            if [ -n "${9}" -a -z "${10}" ]; then
+                macaddr="${9}"
+            elif [ -n "${9}" -a -n "${10}" -a -n "${11}" -a -n "${12}" -a -n "${13}" -a -n "${14}" ]; then
+                macaddr="${9}:${10}:${11}:${12}:${13}:${14}"
+            fi
+            ;;
+    esac
+    return 0
+}
+
+route_to_var() {
+    local v="${1}":
+    local i
+    set --
+    while [ -n "$v" ]; do
+        if [ "${v#\[*:*:*\]:}" != "$v" ]; then
+            # handle IPv6 address
+            i="${v%%\]:*}"
+            i="${i##\[}"
+            set -- "$@" "$i"
+            v=${v#\["$i"\]:}
+        else
+            set -- "$@" "${v%%:*}"
+            v=${v#*:}
+        fi
+    done
+
+    unset route_mask route_gw route_dev
+    case $# in
+        2)
+            [ -n "$1" ] && route_mask="$1"
+            [ -n "$2" ] && route_gw="$2"
+            return 0
+            ;;
+        3)
+            [ -n "$1" ] && route_mask="$1"
+            [ -n "$2" ] && route_gw="$2"
+            [ -n "$3" ] && route_dev="$3"
+            return 0
+            ;;
+        *) return 1 ;;
+    esac
+}
+
+parse_ifname_opts() {
+    local IFS=:
+    # shellcheck disable=SC2086
+    set -- $1
+
+    case $# in
+        7)
+            ifname_if=$1
+            # udev requires MAC addresses to be lower case
+            ifname_mac=$(echo "$2:$3:$4:$5:$6:$7" | sed 'y/ABCDEF/abcdef/')
+            ;;
+        21)
+            ifname_if=$1
+            # udev requires MAC addresses to be lower case
+            ifname_mac=$(echo "$2:$3:$4:$5:$6:$7:$8:$9:${10}:${11}:${12}:${13}:${14}:${15}:${16}:${17}:${18}:${19}:${20}:${21}" | sed 'y/ABCDEF/abcdef/')
+            ;;
+        *)
+            die "Invalid arguments for ifname="
+            ;;
+    esac
+
+    case $ifname_if in
+        eth[0-9] | eth[0-9][0-9] | eth[0-9][0-9][0-9] | eth[0-9][0-9][0-9][0-9])
+            warn "ifname=$ifname_if uses the kernel name space for interfaces"
+            warn "This can fail for multiple network interfaces and is discouraged!"
+            warn 'Please use a custom name like "netboot" or "bluesocket"'
+            warn "or use biosdevname and no ifname= at all."
+            ;;
+    esac
+
+}
+
+# some network driver need long time to initialize, wait before it's ready.
+wait_for_if_link() {
+    local cnt=0
+    local li
+    local timeout
+    timeout=$(getargs rd.net.timeout.iflink=)
+    timeout=${timeout:-60}
+    timeout=$((timeout * 10))
+
+    while [ $cnt -lt $timeout ]; do
+        li=$(ip link show dev "$@" 2> /dev/null)
+        [ -n "$li" ] && return 0
+        sleep 0.1
+        cnt=$((cnt + 1))
+    done
+    return 1
+}
+
+wait_for_if_up() {
+    local cnt=0
+    local li
+    local timeout
+    timeout=$(getargs rd.net.timeout.ifup=)
+    timeout=${timeout:-20}
+    timeout=$((timeout * 10))
+
+    while [ $cnt -lt $timeout ]; do
+        li=$(ip link show up dev "$@")
+        if [ -n "$li" ]; then
+            case "$li" in
+                *\<UP*)
+                    return 0
+                    ;;
+                *\<*,UP\>*)
+                    return 0
+                    ;;
+                *\<*,UP,*\>*)
+                    return 0
+                    ;;
+            esac
+        fi
+        if strstr "$li" "LOWER_UP" \
+            && strstr "$li" "state UNKNOWN" \
+            && ! strstr "$li" "DORMANT"; then
+            return 0
+        fi
+        sleep 0.1
+        cnt=$((cnt + 1))
+    done
+    return 1
+}
+
+wait_for_route_ok() {
+    local cnt=0
+    local timeout
+    timeout=$(getargs rd.net.timeout.route=)
+    timeout=${timeout:-20}
+    timeout=$((timeout * 10))
+
+    while [ $cnt -lt $timeout ]; do
+        li=$(ip route show)
+        [ -n "$li" ] && [ -z "${li##*"$1"*}" ] && return 0
+        sleep 0.1
+        cnt=$((cnt + 1))
+    done
+    return 1
+}
+
+wait_for_ipv6_dad_link() {
+    local cnt=0
+    local timeout
+    timeout=$(getargs rd.net.timeout.ipv6dad=)
+    timeout=${timeout:-50}
+    timeout=$((timeout * 10))
+
+    while [ $cnt -lt $timeout ]; do
+        [ -n "$(ip -6 addr show dev "$@" scope link)" ] \
+            && [ -z "$(ip -6 addr show dev "$@" scope link tentative)" ] \
+            && return 0
+        [ -n "$(ip -6 addr show dev "$@" scope link dadfailed)" ] \
+            && return 1
+        sleep 0.1
+        cnt=$((cnt + 1))
+    done
+    return 1
+}
+
+wait_for_ipv6_dad() {
+    local cnt=0
+    local timeout
+    timeout=$(getargs rd.net.timeout.ipv6dad=)
+    timeout=${timeout:-50}
+    timeout=$((timeout * 10))
+
+    while [ $cnt -lt $timeout ]; do
+        [ -n "$(ip -6 addr show dev "$@")" ] \
+            && [ -z "$(ip -6 addr show dev "$@" tentative)" ] \
+            && return 0
+        [ -n "$(ip -6 addr show dev "$@" dadfailed)" ] \
+            && return 1
+        sleep 0.1
+        cnt=$((cnt + 1))
+    done
+    return 1
+}
+
+wait_for_ipv6_auto() {
+    local cnt=0
+    local timeout
+    timeout=$(getargs rd.net.timeout.ipv6auto=)
+    timeout=${timeout:-40}
+    timeout=$((timeout * 10))
+
+    while [ $cnt -lt $timeout ]; do
+        [ -z "$(ip -6 addr show dev "$@" tentative)" ] \
+            && { ip -6 route list proto ra dev "$@" | grep -q ^default; } \
+            && return 0
+        sleep 0.1
+        cnt=$((cnt + 1))
+    done
+    return 1
+}
+
+linkup() {
+    wait_for_if_link "$@" 2> /dev/null && ip link set "$@" up 2> /dev/null && wait_for_if_up "$@" 2> /dev/null
+}
+
+type hostname > /dev/null 2>&1 \
+    || hostname() {
+        cat /proc/sys/kernel/hostname
+    }
+
+iface_has_carrier() {
+    local cnt=0
+    local iface="$1" flags=""
+    local timeout
+    local iface_sys_path
+    [ -n "$iface" ] || return 2
+    iface_sys_path="/sys/class/net/$iface"
+    [ -d "$iface_sys_path" ] || return 2
+    timeout=$(getargs rd.net.timeout.carrier=)
+    timeout=${timeout:-10}
+    timeout=$((timeout * 10))
+
+    linkup "$1"
+
+    li=$(ip link show up dev "$iface")
+    strstr "$li" "NO-CARRIER" && _no_carrier_flag=1
+
+    while [ $cnt -lt $timeout ]; do
+        if [ -n "$_no_carrier_flag" ]; then
+            li=$(ip link show up dev "$iface")
+            # NO-CARRIER flag was cleared
+            strstr "$li" "NO-CARRIER" || return 0
+        elif ! [ -e "$iface_sys_path/carrier" ]; then
+            # sysfs not available and "NO-CARRIER" not displayed
+            return 0
+        fi
+        # double check the syscfs carrier flag
+        [ -e "$iface_sys_path/carrier" ] && [ "$(cat "$iface_sys_path"/carrier)" = 1 ] && return 0
+        sleep 0.1
+        cnt=$((cnt + 1))
+    done
+    return 1
+}
+
+iface_has_link() {
+    iface_has_carrier "$@"
+}
+
+iface_is_enslaved() {
+    local _li
+    _li=$(ip link show dev "$@")
+    strstr "$_li" " master " || return 1
+    return 0
+}
+
+find_iface_with_link() {
+    local iface_path="" iface=""
+    for iface_path in /sys/class/net/*; do
+        iface=${iface_path##*/}
+        str_starts "$iface" "lo" && continue
+        if iface_has_link "$iface"; then
+            echo "$iface"
+            return 0
+        fi
+    done
+    return 1
+}
+
+is_persistent_ethernet_name() {
+    local _netif="$1"
+    local _name_assign_type="0"
+
+    [ -f "/sys/class/net/$_netif/name_assign_type" ] \
+        && read -r _name_assign_type < "/sys/class/net/$_netif/name_assign_type" 2> /dev/null
+
+    # NET_NAME_ENUM 1
+    [ "$_name_assign_type" = "1" ] && return 1
+
+    # NET_NAME_PREDICTABLE 2
+    [ "$_name_assign_type" = "2" ] && return 0
+
+    case "$_netif" in
+        # udev persistent interface names
+        eno[0-9] | eno[0-9][0-9] | eno[0-9][0-9][0-9]*) ;;
+
+        ens[0-9] | ens[0-9][0-9] | ens[0-9][0-9][0-9]*) ;;
+
+        enp[0-9]s[0-9]* | enp[0-9][0-9]s[0-9]* | enp[0-9][0-9][0-9]*s[0-9]*) ;;
+
+        enP*p[0-9]s[0-9]* | enP*p[0-9][0-9]s[0-9]* | enP*p[0-9][0-9][0-9]*s[0-9]*) ;;
+
+            # biosdevname
+        em[0-9] | em[0-9][0-9] | em[0-9][0-9][0-9]*) ;;
+
+        p[0-9]p[0-9]* | p[0-9][0-9]p[0-9]* | p[0-9][0-9][0-9]*p[0-9]*) ;;
+
+        *)
+            return 1
+            ;;
+    esac
+    return 0
+}
+
+is_kernel_ethernet_name() {
+    local _netif="$1"
+    local _name_assign_type="1"
+
+    if [ -e "/sys/class/net/$_netif/name_assign_type" ]; then
+        read -r _name_assign_type < "/sys/class/net/$_netif/name_assign_type"
+
+        case "$_name_assign_type" in
+            2 | 3 | 4)
+                # NET_NAME_PREDICTABLE 2
+                # NET_NAME_USER 3
+                # NET_NAME_RENAMED 4
+                return 1
+                ;;
+            1 | *)
+                # NET_NAME_ENUM 1
+                return 0
+                ;;
+        esac
+    fi
+
+    # fallback to error prone manual name check
+    case "$_netif" in
+        eth[0-9] | eth[0-9][0-9] | eth[0-9][0-9][0-9]*)
+            return 0
+            ;;
+        *)
+            return 1
+            ;;
+    esac
+
+}
+
+iface_get_subchannels() {
+    local _netif
+    local _subchannels
+
+    _netif="$1"
+
+    _subchannels=$({
+        for i in /sys/class/net/"$_netif"/device/cdev[0-9]*; do
+            [ -e "$i" ] || continue
+            channel=$(readlink -f "$i")
+            printf -- "%s" "${channel##*/},"
+        done
+    })
+    [ -n "$_subchannels" ] || return 1
+
+    printf -- "%s" "${_subchannels%,}"
+}
diff --git a/modules.d/45net-lib/netroot.sh b/modules.d/45net-lib/netroot.sh
new file mode 100755
index 0000000..8f97774
--- /dev/null
+++ b/modules.d/45net-lib/netroot.sh
@@ -0,0 +1,92 @@
+#!/bin/sh
+
+PATH=/usr/sbin:/usr/bin:/sbin:/bin
+command -v getarg > /dev/null || . /lib/dracut-lib.sh
+command -v setup_net > /dev/null || . /lib/net-lib.sh
+
+# Huh? Empty $1?
+[ -z "$1" ] && exit 1
+
+# [ ! -z $2 ] means this is for manually bringing up network
+# instead of real netroot; If It's called without $2, then there's
+# no sense in doing something if no (net)root info is available
+# or root is already there
+[ -d "$NEWROOT"/proc ] && exit 0
+
+if [ -z "$netroot" ]; then
+    netroot=$(getarg netroot=)
+fi
+
+[ -z "$netroot" ] && exit 1
+
+# Set or override primary interface
+netif=$1
+[ -e "/tmp/net.bootdev" ] && read -r netif < /tmp/net.bootdev
+
+case "$netif" in
+    ??:??:??:??:??:??) # MAC address
+        for i in /sys/class/net/*/address; do
+            read -r mac < "$i"
+            if [ "$mac" = "$netif" ]; then
+                i=${i%/address}
+                netif=${i##*/}
+                break
+            fi
+        done ;;
+esac
+
+# Figure out the handler for root=dhcp by recalling all netroot cmdline
+# handlers when this is not called from manually network bringing up.
+if [ -z "$2" ]; then
+    if getarg "root=dhcp" || getarg "netroot=dhcp" || getarg "root=dhcp6" || getarg "netroot=dhcp6"; then
+        # Load dhcp options
+        # shellcheck disable=SC1090
+        [ -e /tmp/dhclient."$netif".dhcpopts ] && . /tmp/dhclient."$netif".dhcpopts
+
+        # If we have a specific bootdev with no dhcpoptions or empty root-path,
+        # we die. Otherwise we just warn
+        if [ -z "$new_root_path" ]; then
+            [ -n "$BOOTDEV" ] && die "No dhcp root-path received for '$BOOTDEV'"
+            warn "No dhcp root-path received for '$netif' trying other interfaces if available"
+            exit 1
+        fi
+
+        rm -f -- "$hookdir"/initqueue/finished/dhcp.sh
+
+        # Set netroot to new_root_path, so cmdline parsers don't call
+        netroot=$new_root_path
+
+        # FIXME!
+        unset rootok
+        for f in "$hookdir"/cmdline/90*.sh; do
+            # shellcheck disable=SC1090
+            [ -f "$f" ] && . "$f"
+        done
+    else
+        rootok="1"
+    fi
+
+    # Check: do we really know how to handle (net)root?
+    if [ -z "$root" ]; then
+        root=$(getarg root=)
+    fi
+    [ -z "$root" ] && die "No or empty root= argument"
+    [ -z "$rootok" ] && die "Don't know how to handle 'root=$root'"
+
+    handler=${netroot%%:*}
+    handler=${handler%%4}
+    handler=$(command -v "${handler}"root)
+    if [ -z "$netroot" ] || [ ! -e "$handler" ]; then
+        die "No handler for netroot type '$netroot'"
+    fi
+fi
+
+# Source netroot hooks before we start the handler
+source_hook netroot "$netif"
+
+# Run the handler; don't store the root, it may change from device to device
+# XXX other variables to export?
+[ -n "$handler" ] && "$handler" "$netif" "$netroot" "$NEWROOT"
+save_netinfo "$netif"
+
+exit 0