Index: stable/10/etc/rc.subr =================================================================== --- stable/10/etc/rc.subr (revision 272867) +++ stable/10/etc/rc.subr (revision 272868) @@ -1,2029 +1,2045 @@ # $NetBSD: rc.subr,v 1.67 2006/10/07 11:25:15 elad Exp $ # $FreeBSD$ # # Copyright (c) 1997-2004 The NetBSD Foundation, Inc. # All rights reserved. # # This code is derived from software contributed to The NetBSD Foundation # by Luke Mewburn. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions # are met: # 1. Redistributions of source code must retain the above copyright # notice, this list of conditions and the following disclaimer. # 2. Redistributions in binary form must reproduce the above copyright # notice, this list of conditions and the following disclaimer in the # documentation and/or other materials provided with the distribution. # # THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS # ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED # TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR # PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS # BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR # CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF # SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS # INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN # CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) # ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE # POSSIBILITY OF SUCH DAMAGE. # # rc.subr # functions used by various rc scripts # : ${RC_PID:=$$}; export RC_PID # # Operating System dependent/independent variables # if [ -z "${_rc_subr_loaded}" ]; then _rc_subr_loaded="YES" SYSCTL="/sbin/sysctl" SYSCTL_N="${SYSCTL} -n" SYSCTL_W="${SYSCTL}" ID="/usr/bin/id" IDCMD="if [ -x $ID ]; then $ID -un; fi" PS="/bin/ps -ww" JID=`$PS -p $$ -o jid=` # # functions # --------- # list_vars pattern # List vars matching pattern. # list_vars() { set | { while read LINE; do var="${LINE%%=*}" case "$var" in "$LINE"|*[!a-zA-Z0-9_]*) continue ;; $1) echo $var esac done; } } # set_rcvar_obsolete oldvar [newvar] [msg] # Define obsolete variable. # Global variable $rcvars_obsolete is used. # set_rcvar_obsolete() { local _var _var=$1 debug "rcvar_obsolete: \$$1(old) -> \$$2(new) is defined" rcvars_obsolete="${rcvars_obsolete# } $1" eval ${1}_newvar=\"$2\" shift 2 eval ${_var}_obsolete_msg=\"$*\" } # # force_depend script [rcvar] # Force a service to start. Intended for use by services # to resolve dependency issues. # $1 - filename of script, in /etc/rc.d, to run # $2 - name of the script's rcvar (minus the _enable) # force_depend() { local _depend _dep_rcvar _depend="$1" _dep_rcvar="${2:-$1}_enable" [ -n "$rc_fast" ] && ! checkyesno always_force_depends && checkyesno $_dep_rcvar && return 0 /etc/rc.d/${_depend} forcestatus >/dev/null 2>&1 && return 0 info "${name} depends on ${_depend}, which will be forced to start." if ! /etc/rc.d/${_depend} forcestart; then warn "Unable to force ${_depend}. It may already be running." return 1 fi } # # checkyesno var # Test $1 variable, and warn if not set to YES or NO. # Return 0 if it's "yes" (et al), nonzero otherwise. # checkyesno() { eval _value=\$${1} debug "checkyesno: $1 is set to $_value." case $_value in # "yes", "true", "on", or "1" [Yy][Ee][Ss]|[Tt][Rr][Uu][Ee]|[Oo][Nn]|1) return 0 ;; # "no", "false", "off", or "0" [Nn][Oo]|[Ff][Aa][Ll][Ss][Ee]|[Oo][Ff][Ff]|0) return 1 ;; *) warn "\$${1} is not set properly - see rc.conf(5)." return 1 ;; esac } # # reverse_list list # print the list in reverse order # reverse_list() { _revlist= for _revfile; do _revlist="$_revfile $_revlist" done echo $_revlist } # stop_boot always # If booting directly to multiuser or $always is enabled, # send SIGTERM to the parent (/etc/rc) to abort the boot. # Otherwise just exit. # stop_boot() { local always case $1 in # "yes", "true", "on", or "1" [Yy][Ee][Ss]|[Tt][Rr][Uu][Ee]|[Oo][Nn]|1) always=true ;; *) always=false ;; esac if [ "$autoboot" = yes -o "$always" = true ]; then echo "ERROR: ABORTING BOOT (sending SIGTERM to parent)!" kill -TERM ${RC_PID} fi exit 1 } # # mount_critical_filesystems type # Go through the list of critical filesystems as provided in # the rc.conf(5) variable $critical_filesystems_${type}, checking # each one to see if it is mounted, and if it is not, mounting it. # mount_critical_filesystems() { eval _fslist=\$critical_filesystems_${1} for _fs in $_fslist; do mount | ( _ismounted=false while read what _on on _type type; do if [ $on = $_fs ]; then _ismounted=true fi done if $_ismounted; then : else mount $_fs >/dev/null 2>&1 fi ) done } # # check_pidfile pidfile procname [interpreter] # Parses the first line of pidfile for a PID, and ensures # that the process is running and matches procname. # Prints the matching PID upon success, nothing otherwise. # interpreter is optional; see _find_processes() for details. # check_pidfile() { _pidfile=$1 _procname=$2 _interpreter=$3 if [ -z "$_pidfile" -o -z "$_procname" ]; then err 3 'USAGE: check_pidfile pidfile procname [interpreter]' fi if [ ! -f $_pidfile ]; then debug "pid file ($_pidfile): not readable." return fi read _pid _junk < $_pidfile if [ -z "$_pid" ]; then debug "pid file ($_pidfile): no pid in file." return fi _find_processes $_procname ${_interpreter:-.} '-p '"$_pid" } # # check_process procname [interpreter] # Ensures that a process (or processes) named procname is running. # Prints a list of matching PIDs. # interpreter is optional; see _find_processes() for details. # check_process() { _procname=$1 _interpreter=$2 if [ -z "$_procname" ]; then err 3 'USAGE: check_process procname [interpreter]' fi _find_processes $_procname ${_interpreter:-.} '-ax' } # # _find_processes procname interpreter psargs # Search for procname in the output of ps generated by psargs. # Prints the PIDs of any matching processes, space separated. # # If interpreter == ".", check the following variations of procname # against the first word of each command: # procname # `basename procname` # `basename procname` + ":" # "(" + `basename procname` + ")" # "[" + `basename procname` + "]" # # If interpreter != ".", read the first line of procname, remove the # leading #!, normalise whitespace, append procname, and attempt to # match that against each command, either as is, or with extra words # at the end. As an alternative, to deal with interpreted daemons # using perl, the basename of the interpreter plus a colon is also # tried as the prefix to procname. # _find_processes() { if [ $# -ne 3 ]; then err 3 'USAGE: _find_processes procname interpreter psargs' fi _procname=$1 _interpreter=$2 _psargs=$3 _pref= if [ $_interpreter != "." ]; then # an interpreted script _script="${_chroot}${_chroot:+/}$_procname" if [ -r "$_script" ]; then read _interp < $_script # read interpreter name case "$_interp" in \#!*) _interp=${_interp#\#!} # strip #! set -- $_interp case $1 in */bin/env) shift # drop env to get real name ;; esac if [ $_interpreter != $1 ]; then warn "\$command_interpreter $_interpreter != $1" fi ;; *) warn "no shebang line in $_script" set -- $_interpreter ;; esac else warn "cannot read shebang line from $_script" set -- $_interpreter fi _interp="$* $_procname" # cleanup spaces, add _procname _interpbn=${1##*/} _fp_args='_argv' _fp_match='case "$_argv" in ${_interp}|"${_interp} "*|"[${_interpbn}]"|"${_interpbn}: ${_procname}"*)' else # a normal daemon _procnamebn=${_procname##*/} _fp_args='_arg0 _argv' _fp_match='case "$_arg0" in $_procname|$_procnamebn|${_procnamebn}:|"(${_procnamebn})"|"[${_procnamebn}]")' fi _proccheck="\ $PS 2>/dev/null -o pid= -o jid= -o command= $_psargs"' | while read _npid _jid '"$_fp_args"'; do '"$_fp_match"' if [ "$JID" -eq "$_jid" ]; then echo -n "$_pref$_npid"; _pref=" "; fi ;; esac done' # debug "in _find_processes: proccheck is ($_proccheck)." eval $_proccheck } # sort_lite [-b] [-n] [-k POS] [-t SEP] # A lite version of sort(1) (supporting a few options) that can be used # before the real sort(1) is available (e.g., in scripts that run prior # to mountcritremote). Requires only shell built-in functionality. # sort_lite() { local funcname=sort_lite local sort_sep="$IFS" sort_ignore_leading_space= local sort_field=0 sort_strict_fields= sort_numeric= local nitems=0 skip_leading=0 trim= local OPTIND flag while getopts bnk:t: flag; do case "$flag" in b) sort_ignore_leading_space=1 ;; n) sort_numeric=1 sort_ignore_leading_space=1 ;; k) sort_field="${OPTARG%%,*}" ;; # only up to first comma # NB: Unlike sort(1) only one POS allowed t) sort_sep="$OPTARG" if [ ${#sort_sep} -gt 1 ]; then echo "$funcname: multi-character tab \`$sort_sep'" >&2 return 1 fi sort_strict_fields=1 ;; \?) return 1 ;; esac done shift $(( $OPTIND - 1 )) # Create transformation pattern to trim leading text if desired case "$sort_field" in ""|[!0-9]*|*[!0-9.]*) echo "$funcname: invalid sort field \`$sort_field'" >&2 return 1 ;; *.*) skip_leading=${sort_field#*.} sort_field=${sort_field%%.*} while [ ${skip_leading:-0} -gt 1 ] 2> /dev/null; do trim="$trim?" skip_leading=$(( $skip_leading - 1 )) done esac # Copy input to series of local numbered variables # NB: IFS of NULL preserves leading whitespace local LINE while IFS= read -r LINE || [ "$LINE" ]; do nitems=$(( $nitems + 1 )) local src_$nitems="$LINE" done # # Sort numbered locals using insertion sort # local curitem curitem_orig curitem_mod curitem_haskey local dest dest_orig dest_mod dest_haskey local d gt n local i=1 while [ $i -le $nitems ]; do curitem_haskey=1 # Assume sort field (-k POS) exists eval curitem=\"\$src_$i\" curitem_mod="$curitem" # for modified comparison curitem_orig="$curitem" # for original comparison # Trim leading whitespace if desired if [ "$sort_ignore_leading_space" ]; then while case "$curitem_orig" in [$IFS]*) : ;; *) false; esac do curitem_orig="${curitem_orig#?}" done curitem_mod="$curitem_orig" fi # Shift modified comparison value if sort field (-k POS) is > 1 n=$sort_field while [ $n -gt 1 ]; do case "$curitem_mod" in *[$sort_sep]*) # Cut text up-to (and incl.) first separator curitem_mod="${curitem_mod#*[$sort_sep]}" # Skip NULLs unless strict field splitting [ "$sort_strict_fields" ] || [ "${curitem_mod%%[$sort_sep]*}" ] || [ $n -eq 2 ] || continue ;; *) # Asked for a field that doesn't exist curitem_haskey= break esac n=$(( $n - 1 )) done # Trim trailing words if sort field >= 1 [ $sort_field -ge 1 -a "$sort_numeric" ] && curitem_mod="${curitem_mod%%[$sort_sep]*}" # Apply optional trim (-k POS.TRIM) to cut leading characters curitem_mod="${curitem_mod#$trim}" # Determine the type of modified comparison to use initially # NB: Prefer numerical if requested but fallback to standard case "$curitem_mod" in ""|[!0-9]*) # NULL or begins with non-number gt=">" [ "$sort_numeric" ] && curitem_mod=0 ;; *) if [ "$sort_numeric" ]; then gt="-gt" curitem_mod="${curitem_mod%%[!0-9]*}" # NB: trailing non-digits removed # otherwise numeric comparison fails else gt=">" fi esac # If first time through, short-circuit below position-search if [ $i -le 1 ]; then d=0 else d=1 fi # # Find appropriate element position # while [ $d -gt 0 ] do dest_haskey=$curitem_haskey eval dest=\"\$dest_$d\" dest_mod="$dest" # for modified comparison dest_orig="$dest" # for original comparison # Trim leading whitespace if desired if [ "$sort_ignore_leading_space" ]; then while case "$dest_orig" in [$IFS]*) : ;; *) false; esac do dest_orig="${dest_orig#?}" done dest_mod="$dest_orig" fi # Shift modified value if sort field (-k POS) is > 1 n=$sort_field while [ $n -gt 1 ]; do case "$dest_mod" in *[$sort_sep]*) # Cut text up-to (and incl.) 1st sep dest_mod="${dest_mod#*[$sort_sep]}" # Skip NULLs unless strict fields [ "$sort_strict_fields" ] || [ "${dest_mod%%[$sort_sep]*}" ] || [ $n -eq 2 ] || continue ;; *) # Asked for a field that doesn't exist dest_haskey= break esac n=$(( $n - 1 )) done # Trim trailing words if sort field >= 1 [ $sort_field -ge 1 -a "$sort_numeric" ] && dest_mod="${dest_mod%%[$sort_sep]*}" # Apply optional trim (-k POS.TRIM), cut leading chars dest_mod="${dest_mod#$trim}" # Determine type of modified comparison to use # NB: Prefer numerical if requested, fallback to std case "$dest_mod" in ""|[!0-9]*) # NULL or begins with non-number gt=">" [ "$sort_numeric" ] && dest_mod=0 ;; *) if [ "$sort_numeric" ]; then gt="-gt" dest_mod="${dest_mod%%[!0-9]*}" # NB: kill trailing non-digits # for numeric comparison safety else gt=">" fi esac # Break if we've found the proper element position if [ "$curitem_haskey" -a "$dest_haskey" ]; then if [ "$dest_mod" = "$curitem_mod" ]; then [ "$dest_orig" ">" "$curitem_orig" ] && break elif [ "$dest_mod" $gt "$curitem_mod" ] \ 2> /dev/null then break fi else [ "$dest_orig" ">" "$curitem_orig" ] && break fi # Break if we've hit the end [ $d -ge $i ] && break d=$(( $d + 1 )) done # Shift remaining positions forward, making room for new item n=$i while [ $n -ge $d ]; do # Shift destination item forward one placement eval dest_$(( $n + 1 ))=\"\$dest_$n\" n=$(( $n - 1 )) done # Place the element if [ $i -eq 1 ]; then local dest_1="$curitem" else local dest_$d="$curitem" fi i=$(( $i + 1 )) done # Print sorted results d=1 while [ $d -le $nitems ]; do eval echo \"\$dest_$d\" d=$(( $d + 1 )) done } # # wait_for_pids pid [pid ...] # spins until none of the pids exist # wait_for_pids() { local _list _prefix _nlist _j _list="$@" if [ -z "$_list" ]; then return fi _prefix= while true; do _nlist=""; for _j in $_list; do if kill -0 $_j 2>/dev/null; then _nlist="${_nlist}${_nlist:+ }$_j" [ -n "$_prefix" ] && sleep 1 fi done if [ -z "$_nlist" ]; then break fi _list=$_nlist echo -n ${_prefix:-"Waiting for PIDS: "}$_list _prefix=", " pwait $_list 2>/dev/null done if [ -n "$_prefix" ]; then echo "." fi } # # get_pidfile_from_conf string file # # Takes a string to search for in the specified file. # Ignores lines with traditional comment characters. # # Example: # # if get_pidfile_from_conf string file; then # pidfile="$_pidfile_from_conf" # else # pidfile='appropriate default' # fi # get_pidfile_from_conf() { if [ -z "$1" -o -z "$2" ]; then err 3 "USAGE: get_pidfile_from_conf string file ($name)" fi local string file line string="$1" ; file="$2" if [ ! -s "$file" ]; then err 3 "get_pidfile_from_conf: $file does not exist ($name)" fi while read line; do case "$line" in *[#\;]*${string}*) continue ;; *${string}*) break ;; esac done < $file if [ -n "$line" ]; then line=${line#*/} _pidfile_from_conf="/${line%%[\"\;]*}" else return 1 fi } # # check_startmsgs # If rc_quiet is set (usually as a result of using faststart at # boot time) check if rc_startmsgs is enabled. # check_startmsgs() { if [ -n "$rc_quiet" ]; then checkyesno rc_startmsgs else return 0 fi } # # run_rc_command argument # Search for argument in the list of supported commands, which is: # "start stop restart rcvar status poll ${extra_commands}" # If there's a match, run ${argument}_cmd or the default method # (see below). # # If argument has a given prefix, then change the operation as follows: # Prefix Operation # ------ --------- # fast Skip the pid check, and set rc_fast=yes, rc_quiet=yes # force Set ${rcvar} to YES, and set rc_force=yes # one Set ${rcvar} to YES # quiet Don't output some diagnostics, and set rc_quiet=yes # # The following globals are used: # # Name Needed Purpose # ---- ------ ------- # name y Name of script. # # command n Full path to command. # Not needed if ${rc_arg}_cmd is set for # each keyword. # # command_args n Optional args/shell directives for command. # # command_interpreter n If not empty, command is interpreted, so # call check_{pidfile,process}() appropriately. # # desc n Description of script. # # extra_commands n List of extra commands supported. # # pidfile n If set, use check_pidfile $pidfile $command, # otherwise use check_process $command. # In either case, only check if $command is set. # # procname n Process name to check for instead of $command. # # rcvar n This is checked with checkyesno to determine # if the action should be run. # # ${name}_program n Full path to command. # Meant to be used in /etc/rc.conf to override # ${command}. # # ${name}_chroot n Directory to chroot to before running ${command} # Requires /usr to be mounted. # # ${name}_chdir n Directory to cd to before running ${command} # (if not using ${name}_chroot). # # ${name}_flags n Arguments to call ${command} with. # NOTE: $flags from the parent environment # can be used to override this. # # ${name}_fib n Routing table number to run ${command} with. # # ${name}_nice n Nice level to run ${command} at. # # ${name}_user n User to run ${command} as, using su(1) if not # using ${name}_chroot. # Requires /usr to be mounted. # # ${name}_group n Group to run chrooted ${command} as. # Requires /usr to be mounted. # # ${name}_groups n Comma separated list of supplementary groups # to run the chrooted ${command} with. # Requires /usr to be mounted. # # ${rc_arg}_cmd n If set, use this as the method when invoked; # Otherwise, use default command (see below) # # ${rc_arg}_precmd n If set, run just before performing the # ${rc_arg}_cmd method in the default # operation (i.e, after checking for required # bits and process (non)existence). # If this completes with a non-zero exit code, # don't run ${rc_arg}_cmd. # # ${rc_arg}_postcmd n If set, run just after performing the # ${rc_arg}_cmd method, if that method # returned a zero exit code. # # required_dirs n If set, check for the existence of the given # directories before running a (re)start command. # # required_files n If set, check for the readability of the given # files before running a (re)start command. # # required_modules n If set, ensure the given kernel modules are # loaded before running a (re)start command. # The check and possible loads are actually # done after start_precmd so that the modules # aren't loaded in vain, should the precmd # return a non-zero status to indicate a error. # If a word in the list looks like "foo:bar", # "foo" is the KLD file name and "bar" is the # module name. If a word looks like "foo~bar", # "foo" is the KLD file name and "bar" is a # egrep(1) pattern matching the module name. # Otherwise the module name is assumed to be # the same as the KLD file name, which is most # common. See load_kld(). # # required_vars n If set, perform checkyesno on each of the # listed variables before running the default # (re)start command. # # Default behaviour for a given argument, if no override method is # provided: # # Argument Default behaviour # -------- ----------------- # start if !running && checkyesno ${rcvar} # ${command} # # stop if ${pidfile} # rc_pid=$(check_pidfile $pidfile $command) # else # rc_pid=$(check_process $command) # kill $sig_stop $rc_pid # wait_for_pids $rc_pid # ($sig_stop defaults to TERM.) # # reload Similar to stop, except use $sig_reload instead, # and doesn't wait_for_pids. # $sig_reload defaults to HUP. # Note that `reload' isn't provided by default, # it should be enabled via $extra_commands. # # restart Run `stop' then `start'. # # status Show if ${command} is running, etc. # # poll Wait for ${command} to exit. # # rcvar Display what rc.conf variable is used (if any). # # enabled Return true if the service is enabled. # # Variables available to methods, and after run_rc_command() has # completed: # # Variable Purpose # -------- ------- # rc_arg Argument to command, after fast/force/one processing # performed # # rc_flags Flags to start the default command with. # Defaults to ${name}_flags, unless overridden # by $flags from the environment. # This variable may be changed by the precmd method. # # rc_pid PID of command (if appropriate) # # rc_fast Not empty if "fast" was provided (q.v.) # # rc_force Not empty if "force" was provided (q.v.) # # rc_quiet Not empty if "quiet" was provided # # run_rc_command() { _return=0 rc_arg=$1 if [ -z "$name" ]; then err 3 'run_rc_command: $name is not set.' fi # Don't repeat the first argument when passing additional command- # line arguments to the command subroutines. # shift 1 rc_extra_args="$*" _rc_prefix= case "$rc_arg" in fast*) # "fast" prefix; don't check pid rc_arg=${rc_arg#fast} rc_fast=yes rc_quiet=yes ;; force*) # "force" prefix; always run rc_force=yes _rc_prefix=force rc_arg=${rc_arg#${_rc_prefix}} if [ -n "${rcvar}" ]; then eval ${rcvar}=YES fi ;; one*) # "one" prefix; set ${rcvar}=yes _rc_prefix=one rc_arg=${rc_arg#${_rc_prefix}} if [ -n "${rcvar}" ]; then eval ${rcvar}=YES fi ;; quiet*) # "quiet" prefix; omit some messages _rc_prefix=quiet rc_arg=${rc_arg#${_rc_prefix}} rc_quiet=yes ;; esac eval _override_command=\$${name}_program command=${_override_command:-$command} _keywords="start stop restart rcvar enabled $extra_commands" rc_pid= _pidcmd= _procname=${procname:-${command}} # setup pid check command if [ -n "$_procname" ]; then if [ -n "$pidfile" ]; then _pidcmd='rc_pid=$(check_pidfile '"$pidfile $_procname $command_interpreter"')' else _pidcmd='rc_pid=$(check_process '"$_procname $command_interpreter"')' fi if [ -n "$_pidcmd" ]; then _keywords="${_keywords} status poll" fi fi if [ -z "$rc_arg" ]; then rc_usage $_keywords fi if [ "$rc_arg" = "enabled" ] ; then checkyesno ${rcvar} return $? fi if [ -n "$flags" ]; then # allow override from environment rc_flags=$flags else eval rc_flags=\$${name}_flags fi eval _chdir=\$${name}_chdir _chroot=\$${name}_chroot \ _nice=\$${name}_nice _user=\$${name}_user \ _group=\$${name}_group _groups=\$${name}_groups \ _fib=\$${name}_fib if [ -n "$_user" ]; then # unset $_user if running as that user if [ "$_user" = "$(eval $IDCMD)" ]; then unset _user fi fi [ -z "$autoboot" ] && eval $_pidcmd # determine the pid if necessary for _elem in $_keywords; do if [ "$_elem" != "$rc_arg" ]; then continue fi # if ${rcvar} is set, $1 is not "rcvar" # and ${rc_pid} is not set, then run # checkyesno ${rcvar} # and return if that failed # if [ -n "${rcvar}" -a "$rc_arg" != "rcvar" -a "$rc_arg" != "stop" ] || [ -n "${rcvar}" -a "$rc_arg" = "stop" -a -z "${rc_pid}" ]; then if ! checkyesno ${rcvar}; then if [ -n "${rc_quiet}" ]; then return 0 fi echo -n "Cannot '${rc_arg}' $name. Set ${rcvar} to " echo -n "YES in /etc/rc.conf or use 'one${rc_arg}' " echo "instead of '${rc_arg}'." return 0 fi fi # if there's a custom ${XXX_cmd}, # run that instead of the default # eval _cmd=\$${rc_arg}_cmd \ _precmd=\$${rc_arg}_precmd \ _postcmd=\$${rc_arg}_postcmd if [ -n "$_cmd" ]; then _run_rc_precmd || return 1 _run_rc_doit "$_cmd $rc_extra_args" || return 1 _run_rc_postcmd return $_return fi case "$rc_arg" in # default operations... status) _run_rc_precmd || return 1 if [ -n "$rc_pid" ]; then echo "${name} is running as pid $rc_pid." else echo "${name} is not running." return 1 fi _run_rc_postcmd ;; start) if [ -z "$rc_fast" -a -n "$rc_pid" ]; then if [ -z "$rc_quiet" ]; then echo 1>&2 "${name} already running? " \ "(pid=$rc_pid)." fi return 1 fi if [ ! -x "${_chroot}${_chroot:+/}${command}" ]; then warn "run_rc_command: cannot run $command" return 1 fi if ! _run_rc_precmd; then warn "failed precmd routine for ${name}" return 1 fi # setup the full command to run # check_startmsgs && echo "Starting ${name}." if [ -n "$_chroot" ]; then _doit="\ ${_nice:+nice -n $_nice }\ ${_fib:+setfib -F $_fib }\ chroot ${_user:+-u $_user }${_group:+-g $_group }${_groups:+-G $_groups }\ $_chroot $command $rc_flags $command_args" else _doit="\ ${_chdir:+cd $_chdir && }\ ${_fib:+setfib -F $_fib }\ $command $rc_flags $command_args" if [ -n "$_user" ]; then _doit="su -m $_user -c 'sh -c \"$_doit\"'" fi if [ -n "$_nice" ]; then if [ -z "$_user" ]; then _doit="sh -c \"$_doit\"" fi _doit="nice -n $_nice $_doit" fi fi # run the full command # if ! _run_rc_doit "$_doit"; then warn "failed to start ${name}" return 1 fi # finally, run postcmd # _run_rc_postcmd ;; stop) if [ -z "$rc_pid" ]; then [ -n "$rc_fast" ] && return 0 _run_rc_notrunning return 1 fi _run_rc_precmd || return 1 # send the signal to stop # echo "Stopping ${name}." _doit=$(_run_rc_killcmd "${sig_stop:-TERM}") _run_rc_doit "$_doit" || return 1 # wait for the command to exit, # and run postcmd. wait_for_pids $rc_pid _run_rc_postcmd ;; reload) if [ -z "$rc_pid" ]; then _run_rc_notrunning return 1 fi _run_rc_precmd || return 1 _doit=$(_run_rc_killcmd "${sig_reload:-HUP}") _run_rc_doit "$_doit" || return 1 _run_rc_postcmd ;; restart) # prevent restart being called more # than once by any given script # if ${_rc_restart_done:-false}; then return 0 fi _rc_restart_done=true _run_rc_precmd || return 1 # run those in a subshell to keep global variables ( run_rc_command ${_rc_prefix}stop $rc_extra_args ) ( run_rc_command ${_rc_prefix}start $rc_extra_args ) _return=$? [ $_return -ne 0 ] && [ -z "$rc_force" ] && return 1 _run_rc_postcmd ;; poll) _run_rc_precmd || return 1 if [ -n "$rc_pid" ]; then wait_for_pids $rc_pid fi _run_rc_postcmd ;; rcvar) echo -n "# $name" if [ -n "$desc" ]; then echo " : $desc" else echo "" fi echo "#" # Get unique vars in $rcvar for _v in $rcvar; do case $v in $_v\ *|\ *$_v|*\ $_v\ *) ;; *) v="${v# } $_v" ;; esac done # Display variables. for _v in $v; do if [ -z "$_v" ]; then continue fi eval _desc=\$${_v}_desc eval _defval=\$${_v}_defval _h="-" eval echo \"$_v=\\\"\$$_v\\\"\" # decode multiple lines of _desc while [ -n "$_desc" ]; do case $_desc in *^^*) echo "# $_h ${_desc%%^^*}" _desc=${_desc#*^^} _h=" " ;; *) echo "# $_h ${_desc}" break ;; esac done echo "# (default: \"$_defval\")" done echo "" ;; *) rc_usage $_keywords ;; esac return $_return done echo 1>&2 "$0: unknown directive '$rc_arg'." rc_usage $_keywords # not reached } # # Helper functions for run_rc_command: common code. # They use such global variables besides the exported rc_* ones: # # name R/W # ------------------ # _precmd R # _postcmd R # _return W # _run_rc_precmd() { check_required_before "$rc_arg" || return 1 if [ -n "$_precmd" ]; then debug "run_rc_command: ${rc_arg}_precmd: $_precmd $rc_extra_args" eval "$_precmd $rc_extra_args" _return=$? # If precmd failed and force isn't set, request exit. if [ $_return -ne 0 ] && [ -z "$rc_force" ]; then return 1 fi fi check_required_after "$rc_arg" || return 1 return 0 } _run_rc_postcmd() { if [ -n "$_postcmd" ]; then debug "run_rc_command: ${rc_arg}_postcmd: $_postcmd $rc_extra_args" eval "$_postcmd $rc_extra_args" _return=$? fi return 0 } _run_rc_doit() { debug "run_rc_command: doit: $*" eval "$@" _return=$? # If command failed and force isn't set, request exit. if [ $_return -ne 0 ] && [ -z "$rc_force" ]; then return 1 fi return 0 } _run_rc_notrunning() { local _pidmsg if [ -n "$pidfile" ]; then _pidmsg=" (check $pidfile)." else _pidmsg= fi echo 1>&2 "${name} not running?${_pidmsg}" } _run_rc_killcmd() { local _cmd _cmd="kill -$1 $rc_pid" if [ -n "$_user" ]; then _cmd="su -m ${_user} -c 'sh -c \"${_cmd}\"'" fi echo "$_cmd" } # # run_rc_script file arg # Start the script `file' with `arg', and correctly handle the # return value from the script. # If `file' ends with `.sh', it's sourced into the current environment # when $rc_fast_and_loose is set, otherwise it is run as a child process. # If `file' appears to be a backup or scratch file, ignore it. # Otherwise if it is executable run as a child process. # run_rc_script() { _file=$1 _arg=$2 if [ -z "$_file" -o -z "$_arg" ]; then err 3 'USAGE: run_rc_script file arg' fi unset name command command_args command_interpreter \ extra_commands pidfile procname \ rcvar rcvars_obsolete required_dirs required_files \ required_vars eval unset ${_arg}_cmd ${_arg}_precmd ${_arg}_postcmd case "$_file" in /etc/rc.d/*.sh) # no longer allowed in the base warn "Ignoring old-style startup script $_file" ;; *[~#]|*.OLD|*.bak|*.orig|*,v) # scratch file; skip warn "Ignoring scratch file $_file" ;; *) # run in subshell if [ -x $_file ]; then if [ -n "$rc_fast_and_loose" ]; then set $_arg; . $_file else ( trap "echo Script $_file interrupted >&2 ; kill -QUIT $$" 3 trap "echo Script $_file interrupted >&2 ; exit 1" 2 trap "echo Script $_file running >&2" 29 set $_arg; . $_file ) fi fi ;; esac } # # load_rc_config name # Source in the configuration file for a given name. # load_rc_config() { local _name _rcvar_val _var _defval _v _msg _new _d _name=$1 if [ -z "$_name" ]; then err 3 'USAGE: load_rc_config name' fi if ${_rc_conf_loaded:-false}; then : else if [ -r /etc/defaults/rc.conf ]; then debug "Sourcing /etc/defaults/rc.conf" . /etc/defaults/rc.conf source_rc_confs elif [ -r /etc/rc.conf ]; then debug "Sourcing /etc/rc.conf (/etc/defaults/rc.conf doesn't exist)." . /etc/rc.conf fi _rc_conf_loaded=true fi for _d in /etc ${local_startup%*/rc.d}; do if [ -f ${_d}/rc.conf.d/"$_name" ]; then debug "Sourcing ${_d}/rc.conf.d/$_name" . ${_d}/rc.conf.d/"$_name" elif [ -d ${_d}/rc.conf.d/"$_name" ] ; then local _rc for _rc in ${_d}/rc.conf.d/"$_name"/* ; do if [ -f "$_rc" ] ; then debug "Sourcing $_rc" . "$_rc" fi done fi done # Set defaults if defined. for _var in $rcvar; do eval _defval=\$${_var}_defval if [ -n "$_defval" ]; then eval : \${$_var:=\$${_var}_defval} fi done # check obsolete rc.conf variables for _var in $rcvars_obsolete; do eval _v=\$$_var eval _msg=\$${_var}_obsolete_msg eval _new=\$${_var}_newvar case $_v in "") ;; *) if [ -z "$_new" ]; then _msg="Ignored." else eval $_new=\"\$$_var\" if [ -z "$_msg" ]; then _msg="Use \$$_new instead." fi fi warn "\$$_var is obsolete. $_msg" ;; esac done } # # load_rc_config_var name var # Read the rc.conf(5) var for name and set in the # current shell, using load_rc_config in a subshell to prevent # unwanted side effects from other variable assignments. # load_rc_config_var() { if [ $# -ne 2 ]; then err 3 'USAGE: load_rc_config_var name var' fi eval $(eval '( load_rc_config '$1' >/dev/null; if [ -n "${'$2'}" -o "${'$2'-UNSET}" != "UNSET" ]; then echo '$2'=\'\''${'$2'}\'\''; fi )' ) } # # rc_usage commands # Print a usage string for $0, with `commands' being a list of # valid commands. # rc_usage() { echo -n 1>&2 "Usage: $0 [fast|force|one|quiet](" _sep= for _elem; do echo -n 1>&2 "$_sep$_elem" _sep="|" done echo 1>&2 ")" exit 1 } # # err exitval message # Display message to stderr and log to the syslog, and exit with exitval. # err() { exitval=$1 shift if [ -x /usr/bin/logger ]; then logger "$0: ERROR: $*" fi echo 1>&2 "$0: ERROR: $*" exit $exitval } # # warn message # Display message to stderr and log to the syslog. # warn() { if [ -x /usr/bin/logger ]; then logger "$0: WARNING: $*" fi echo 1>&2 "$0: WARNING: $*" } # # info message # Display informational message to stdout and log to syslog. # info() { case ${rc_info} in [Yy][Ee][Ss]|[Tt][Rr][Uu][Ee]|[Oo][Nn]|1) if [ -x /usr/bin/logger ]; then logger "$0: INFO: $*" fi echo "$0: INFO: $*" ;; esac } # # debug message # If debugging is enabled in rc.conf output message to stderr. # BEWARE that you don't call any subroutine that itself calls this # function. # debug() { case ${rc_debug} in [Yy][Ee][Ss]|[Tt][Rr][Uu][Ee]|[Oo][Nn]|1) if [ -x /usr/bin/logger ]; then logger "$0: DEBUG: $*" fi echo 1>&2 "$0: DEBUG: $*" ;; esac } # # backup_file action file cur backup # Make a backup copy of `file' into `cur', and save the previous # version of `cur' as `backup' or use rcs for archiving. # # This routine checks the value of the backup_uses_rcs variable, # which can be either YES or NO. # # The `action' keyword can be one of the following: # # add `file' is now being backed up (and is possibly # being reentered into the backups system). `cur' # is created and RCS files, if necessary, are # created as well. # # update `file' has changed and needs to be backed up. # If `cur' exists, it is copied to to `back' or # checked into RCS (if the repository file is old), # and then `file' is copied to `cur'. Another RCS # check in done here if RCS is being used. # # remove `file' is no longer being tracked by the backups # system. If RCS is not being used, `cur' is moved # to `back', otherwise an empty file is checked in, # and then `cur' is removed. # # backup_file() { _action=$1 _file=$2 _cur=$3 _back=$4 if checkyesno backup_uses_rcs; then _msg0="backup archive" _msg1="update" # ensure that history file is not locked if [ -f $_cur,v ]; then rcs -q -u -U -M $_cur fi # ensure after switching to rcs that the # current backup is not lost if [ -f $_cur ]; then # no archive, or current newer than archive if [ ! -f $_cur,v -o $_cur -nt $_cur,v ]; then ci -q -f -u -t-"$_msg0" -m"$_msg1" $_cur rcs -q -kb -U $_cur co -q -f -u $_cur fi fi case $_action in add|update) cp -p $_file $_cur ci -q -f -u -t-"$_msg0" -m"$_msg1" $_cur rcs -q -kb -U $_cur co -q -f -u $_cur chown root:wheel $_cur $_cur,v ;; remove) cp /dev/null $_cur ci -q -f -u -t-"$_msg0" -m"$_msg1" $_cur rcs -q -kb -U $_cur chown root:wheel $_cur $_cur,v rm $_cur ;; esac else case $_action in add|update) if [ -f $_cur ]; then cp -p $_cur $_back fi cp -p $_file $_cur chown root:wheel $_cur ;; remove) mv -f $_cur $_back ;; esac fi } # make_symlink src link # Make a symbolic link 'link' to src from basedir. If the # directory in which link is to be created does not exist # a warning will be displayed and an error will be returned. # Returns 0 on success, 1 otherwise. # make_symlink() { local src link linkdir _me src="$1" link="$2" linkdir="`dirname $link`" _me="make_symlink()" if [ -z "$src" -o -z "$link" ]; then warn "$_me: requires two arguments." return 1 fi if [ ! -d "$linkdir" ]; then warn "$_me: the directory $linkdir does not exist." return 1 fi if ! ln -sf $src $link; then warn "$_me: unable to make a symbolic link from $link to $src" return 1 fi return 0 } # devfs_rulesets_from_file file # Reads a set of devfs commands from file, and creates # the specified rulesets with their rules. Returns non-zero # if there was an error. # devfs_rulesets_from_file() { local file _err _me _opts file="$1" _me="devfs_rulesets_from_file" _err=0 if [ -z "$file" ]; then warn "$_me: you must specify a file" return 1 fi if [ ! -e "$file" ]; then debug "$_me: no such file ($file)" return 0 fi # Disable globbing so that the rule patterns are not expanded # by accident with matching filesystem entries. _opts=$-; set -f debug "reading rulesets from file ($file)" { while read line do case $line in \#*) continue ;; \[*\]*) rulenum=`expr "$line" : "\[.*=\([0-9]*\)\]"` if [ -z "$rulenum" ]; then warn "$_me: cannot extract rule number ($line)" _err=1 break fi rulename=`expr "$line" : "\[\(.*\)=[0-9]*\]"` if [ -z "$rulename" ]; then warn "$_me: cannot extract rule name ($line)" _err=1 break; fi eval $rulename=\$rulenum debug "found ruleset: $rulename=$rulenum" if ! /sbin/devfs rule -s $rulenum delset; then _err=1 break fi ;; *) rulecmd="${line%%"\#*"}" # evaluate the command incase it includes # other rules if [ -n "$rulecmd" ]; then debug "adding rule ($rulecmd)" if ! eval /sbin/devfs rule -s $rulenum $rulecmd then _err=1 break fi fi ;; esac if [ $_err -ne 0 ]; then debug "error in $_me" break fi done } < $file case $_opts in *f*) ;; *) set +f ;; esac return $_err } # devfs_init_rulesets # Initializes rulesets from configuration files. Returns # non-zero if there was an error. # devfs_init_rulesets() { local file _me _me="devfs_init_rulesets" # Go through this only once if [ -n "$devfs_rulesets_init" ]; then debug "$_me: devfs rulesets already initialized" return fi for file in $devfs_rulesets; do if ! devfs_rulesets_from_file $file; then warn "$_me: could not read rules from $file" return 1 fi done devfs_rulesets_init=1 debug "$_me: devfs rulesets initialized" return 0 } # devfs_set_ruleset ruleset [dir] # Sets the default ruleset of dir to ruleset. The ruleset argument # must be a ruleset name as specified in devfs.rules(5) file. # Returns non-zero if it could not set it successfully. # devfs_set_ruleset() { local devdir rs _me [ -n "$1" ] && eval rs=\$$1 || rs= [ -n "$2" ] && devdir="-m "$2"" || devdir= _me="devfs_set_ruleset" if [ -z "$rs" ]; then warn "$_me: you must specify a ruleset number" return 1 fi debug "$_me: setting ruleset ($rs) on mount-point (${devdir#-m })" if ! /sbin/devfs $devdir ruleset $rs; then warn "$_me: unable to set ruleset $rs to ${devdir#-m }" return 1 fi return 0 } # devfs_apply_ruleset ruleset [dir] # Apply ruleset number $ruleset to the devfs mountpoint $dir. # The ruleset argument must be a ruleset name as specified # in a devfs.rules(5) file. Returns 0 on success or non-zero # if it could not apply the ruleset. # devfs_apply_ruleset() { local devdir rs _me [ -n "$1" ] && eval rs=\$$1 || rs= [ -n "$2" ] && devdir="-m "$2"" || devdir= _me="devfs_apply_ruleset" if [ -z "$rs" ]; then warn "$_me: you must specify a ruleset" return 1 fi debug "$_me: applying ruleset ($rs) to mount-point (${devdir#-m })" if ! /sbin/devfs $devdir rule -s $rs applyset; then warn "$_me: unable to apply ruleset $rs to ${devdir#-m }" return 1 fi return 0 } # devfs_domount dir [ruleset] # Mount devfs on dir. If ruleset is specified it is set # on the mount-point. It must also be a ruleset name as specified # in a devfs.rules(5) file. Returns 0 on success. # devfs_domount() { local devdir rs _me devdir="$1" [ -n "$2" ] && rs=$2 || rs= _me="devfs_domount()" if [ -z "$devdir" ]; then warn "$_me: you must specify a mount-point" return 1 fi debug "$_me: mount-point is ($devdir), ruleset is ($rs)" if ! mount -t devfs dev "$devdir"; then warn "$_me: Unable to mount devfs on $devdir" return 1 fi if [ -n "$rs" ]; then devfs_init_rulesets devfs_set_ruleset $rs $devdir devfs -m $devdir rule applyset fi return 0 } # Provide a function for normalizing the mounting of memory # filesystems. This should allow the rest of the code here to remain # as close as possible between 5-current and 4-stable. # $1 = size # $2 = mount point # $3 = (optional) extra mdmfs flags mount_md() { if [ -n "$3" ]; then flags="$3" fi /sbin/mdmfs $flags -s $1 md $2 } # Code common to scripts that need to load a kernel module # if it isn't in the kernel yet. Syntax: # load_kld [-e regex] [-m module] file # where -e or -m chooses the way to check if the module # is already loaded: # regex is egrep'd in the output from `kldstat -v', # module is passed to `kldstat -m'. # The default way is as though `-m file' were specified. load_kld() { local _loaded _mod _opt _re while getopts "e:m:" _opt; do case "$_opt" in e) _re="$OPTARG" ;; m) _mod="$OPTARG" ;; *) err 3 'USAGE: load_kld [-e regex] [-m module] file' ;; esac done shift $(($OPTIND - 1)) if [ $# -ne 1 ]; then err 3 'USAGE: load_kld [-e regex] [-m module] file' fi _mod=${_mod:-$1} _loaded=false if [ -n "$_re" ]; then if kldstat -v | egrep -q -e "$_re"; then _loaded=true fi else if kldstat -q -m "$_mod"; then _loaded=true fi fi if ! $_loaded; then if ! kldload "$1"; then warn "Unable to load kernel module $1" return 1 else info "$1 kernel module loaded." fi else debug "load_kld: $1 kernel module already loaded." fi return 0 } # ltr str src dst [var] # Change every $src in $str to $dst. # Useful when /usr is not yet mounted and we cannot use tr(1), sed(1) nor # awk(1). If var is non-NULL, set it to the result. ltr() { local _str _src _dst _out _com _var _str="$1" _src="$2" _dst="$3" _var="$4" _out="" local IFS="${_src}" for _com in ${_str}; do if [ -z "${_out}" ]; then _out="${_com}" else _out="${_out}${_dst}${_com}" fi done if [ -n "${_var}" ]; then setvar "${_var}" "${_out}" else echo "${_out}" fi } # Creates a list of providers for GELI encryption. geli_make_list() { local devices devices2 local provider mountpoint type options rest # Create list of GELI providers from fstab. while read provider mountpoint type options rest ; do case ":${options}" in :*noauto*) noauto=yes ;; *) noauto=no ;; esac case ":${provider}" in :#*) continue ;; *.eli) # Skip swap devices. if [ "${type}" = "swap" -o "${options}" = "sw" -o "${noauto}" = "yes" ]; then continue fi devices="${devices} ${provider}" ;; esac done < /etc/fstab # Append providers from geli_devices. devices="${devices} ${geli_devices}" for provider in ${devices}; do provider=${provider%.eli} provider=${provider#/dev/} devices2="${devices2} ${provider}" done echo ${devices2} } # Find scripts in local_startup directories that use the old syntax # find_local_scripts_old() { zlist='' slist='' for dir in ${local_startup}; do if [ -d "${dir}" ]; then for file in ${dir}/[0-9]*.sh; do grep '^# PROVIDE:' $file >/dev/null 2>&1 && continue zlist="$zlist $file" done for file in ${dir}/[!0-9]*.sh; do grep '^# PROVIDE:' $file >/dev/null 2>&1 && continue slist="$slist $file" done fi done } find_local_scripts_new() { local_rc='' for dir in ${local_startup}; do if [ -d "${dir}" ]; then for file in `grep -l '^# PROVIDE:' ${dir}/* 2>/dev/null`; do case "$file" in *.sample) ;; *) if [ -x "$file" ]; then local_rc="${local_rc} ${file}" fi ;; esac done fi done } # check_required_{before|after} command # Check for things required by the command before and after its precmd, # respectively. The two separate functions are needed because some # conditions should prevent precmd from being run while other things # depend on precmd having already been run. # check_required_before() { local _f case "$1" in start) for _f in $required_vars; do if ! checkyesno $_f; then warn "\$${_f} is not enabled." if [ -z "$rc_force" ]; then return 1 fi fi done for _f in $required_dirs; do if [ ! -d "${_f}/." ]; then warn "${_f} is not a directory." if [ -z "$rc_force" ]; then return 1 fi fi done for _f in $required_files; do if [ ! -r "${_f}" ]; then warn "${_f} is not readable." if [ -z "$rc_force" ]; then return 1 fi fi done ;; esac return 0 } check_required_after() { local _f _args case "$1" in start) for _f in $required_modules; do case "${_f}" in *~*) _args="-e ${_f#*~} ${_f%%~*}" ;; *:*) _args="-m ${_f#*:} ${_f%%:*}" ;; *) _args="${_f}" ;; esac if ! load_kld ${_args}; then if [ -z "$rc_force" ]; then return 1 fi fi done ;; esac return 0 } +# check_jail mib +# Return true if security.jail.$mib exists and set to 1. + +check_jail() +{ + local _mib _v + + _mib=$1 + if _v=$(${SYSCTL_N} "security.jail.$_mib" 2> /dev/null); then + case $_v in + 1) return 0;; + esac + fi + return 1 +} + # check_kern_features mib # Return existence of kern.features.* sysctl MIB as true or # false. The result will be cached in $_rc_cache_kern_features_ # namespace. "0" means the kern.features.X exists. check_kern_features() { local _v [ -n "$1" ] || return 1; eval _v=\$_rc_cache_kern_features_$1 [ -n "$_v" ] && return "$_v"; if ${SYSCTL_N} kern.features.$1 > /dev/null 2>&1; then eval _rc_cache_kern_features_$1=0 return 0 else eval _rc_cache_kern_features_$1=1 return 1 fi } # check_namevarlist var # Return "0" if ${name}_var is reserved in rc.subr. _rc_namevarlist="program chroot chdir flags fib nice user group groups" check_namevarlist() { local _v for _v in $_rc_namevarlist; do case $1 in $_v) return 0 ;; esac done return 1 } # _echoonce var msg mode # mode=0: Echo $msg if ${$var} is empty. # After doing echo, a string is set to ${$var}. # # mode=1: Echo $msg if ${$var} is a string with non-zero length. # _echoonce() { local _var _msg _mode eval _var=\$$1 _msg=$2 _mode=$3 case $_mode in 1) [ -n "$_var" ] && echo "$_msg" ;; *) [ -z "$_var" ] && echo -n "$_msg" && eval "$1=finished" ;; esac } fi # [ -z "${_rc_subr_loaded}" ] _rc_subr_loaded=: Index: stable/10/sys/netinet/ip_fastfwd.c =================================================================== --- stable/10/sys/netinet/ip_fastfwd.c (revision 272867) +++ stable/10/sys/netinet/ip_fastfwd.c (revision 272868) @@ -1,601 +1,601 @@ /*- * Copyright (c) 2003 Andre Oppermann, Internet Business Solutions AG * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. The name of the author may not be used to endorse or promote * products derived from this software without specific prior written * permission. * * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. */ /* * ip_fastforward gets its speed from processing the forwarded packet to * completion (if_output on the other side) without any queues or netisr's. * The receiving interface DMAs the packet into memory, the upper half of * driver calls ip_fastforward, we do our routing table lookup and directly * send it off to the outgoing interface, which DMAs the packet to the * network card. The only part of the packet we touch with the CPU is the * IP header (unless there are complex firewall rules touching other parts * of the packet, but that is up to you). We are essentially limited by bus * bandwidth and how fast the network card/driver can set up receives and * transmits. * * We handle basic errors, IP header errors, checksum errors, * destination unreachable, fragmentation and fragmentation needed and * report them via ICMP to the sender. * * Else if something is not pure IPv4 unicast forwarding we fall back to * the normal ip_input processing path. We should only be called from * interfaces connected to the outside world. * * Firewalling is fully supported including divert, ipfw fwd and ipfilter * ipnat and address rewrite. * * IPSEC is not supported if this host is a tunnel broker. IPSEC is * supported for connections to/from local host. * * We try to do the least expensive (in CPU ops) checks and operations * first to catch junk with as little overhead as possible. * * We take full advantage of hardware support for IP checksum and * fragmentation offloading. * * We don't do ICMP redirect in the fast forwarding path. I have had my own * cases where two core routers with Zebra routing suite would send millions * ICMP redirects to connected hosts if the destination router was not the * default gateway. In one case it was filling the routing table of a host * with approximately 300.000 cloned redirect entries until it ran out of * kernel memory. However the networking code proved very robust and it didn't * crash or fail in other ways. */ /* * Many thanks to Matt Thomas of NetBSD for basic structure of ip_flow.c which * is being followed here. */ #include __FBSDID("$FreeBSD$"); #include "opt_ipfw.h" #include "opt_ipstealth.h" #include "opt_kdtrace.h" #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include static VNET_DEFINE(int, ipfastforward_active); #define V_ipfastforward_active VNET(ipfastforward_active) SYSCTL_VNET_INT(_net_inet_ip, OID_AUTO, fastforwarding, CTLFLAG_RW, &VNET_NAME(ipfastforward_active), 0, "Enable fast IP forwarding"); static struct sockaddr_in * ip_findroute(struct route *ro, struct in_addr dest, struct mbuf *m) { struct sockaddr_in *dst; struct rtentry *rt; /* * Find route to destination. */ bzero(ro, sizeof(*ro)); dst = (struct sockaddr_in *)&ro->ro_dst; dst->sin_family = AF_INET; dst->sin_len = sizeof(*dst); dst->sin_addr.s_addr = dest.s_addr; in_rtalloc_ign(ro, 0, M_GETFIB(m)); /* * Route there and interface still up? */ rt = ro->ro_rt; if (rt && (rt->rt_flags & RTF_UP) && (rt->rt_ifp->if_flags & IFF_UP) && (rt->rt_ifp->if_drv_flags & IFF_DRV_RUNNING)) { if (rt->rt_flags & RTF_GATEWAY) dst = (struct sockaddr_in *)rt->rt_gateway; } else { IPSTAT_INC(ips_noroute); IPSTAT_INC(ips_cantforward); if (rt) RTFREE(rt); icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_HOST, 0, 0); return NULL; } return dst; } /* * Try to forward a packet based on the destination address. * This is a fast path optimized for the plain forwarding case. * If the packet is handled (and consumed) here then we return NULL; * otherwise mbuf is returned and the packet should be delivered * to ip_input for full processing. */ struct mbuf * ip_fastforward(struct mbuf *m) { struct ip *ip; struct mbuf *m0 = NULL; struct route ro; struct sockaddr_in *dst = NULL; struct ifnet *ifp; struct in_addr odest, dest; uint16_t sum, ip_len, ip_off; int error = 0; int hlen, mtu; struct m_tag *fwd_tag = NULL; /* * Are we active and forwarding packets? */ if (!V_ipfastforward_active || !V_ipforwarding) return m; M_ASSERTVALID(m); M_ASSERTPKTHDR(m); bzero(&ro, sizeof(ro)); /* * Step 1: check for packet drop conditions (and sanity checks) */ /* * Is entire packet big enough? */ if (m->m_pkthdr.len < sizeof(struct ip)) { IPSTAT_INC(ips_tooshort); goto drop; } /* * Is first mbuf large enough for ip header and is header present? */ if (m->m_len < sizeof (struct ip) && (m = m_pullup(m, sizeof (struct ip))) == NULL) { IPSTAT_INC(ips_toosmall); return NULL; /* mbuf already free'd */ } ip = mtod(m, struct ip *); /* * Is it IPv4? */ if (ip->ip_v != IPVERSION) { IPSTAT_INC(ips_badvers); goto drop; } /* * Is IP header length correct and is it in first mbuf? */ hlen = ip->ip_hl << 2; if (hlen < sizeof(struct ip)) { /* minimum header length */ IPSTAT_INC(ips_badhlen); goto drop; } if (hlen > m->m_len) { if ((m = m_pullup(m, hlen)) == NULL) { IPSTAT_INC(ips_badhlen); return NULL; /* mbuf already free'd */ } ip = mtod(m, struct ip *); } /* * Checksum correct? */ if (m->m_pkthdr.csum_flags & CSUM_IP_CHECKED) sum = !(m->m_pkthdr.csum_flags & CSUM_IP_VALID); else { if (hlen == sizeof(struct ip)) sum = in_cksum_hdr(ip); else sum = in_cksum(m, hlen); } if (sum) { IPSTAT_INC(ips_badsum); goto drop; } /* * Remember that we have checked the IP header and found it valid. */ m->m_pkthdr.csum_flags |= (CSUM_IP_CHECKED | CSUM_IP_VALID); ip_len = ntohs(ip->ip_len); /* * Is IP length longer than packet we have got? */ if (m->m_pkthdr.len < ip_len) { IPSTAT_INC(ips_tooshort); goto drop; } /* * Is packet longer than IP header tells us? If yes, truncate packet. */ if (m->m_pkthdr.len > ip_len) { if (m->m_len == m->m_pkthdr.len) { m->m_len = ip_len; m->m_pkthdr.len = ip_len; } else m_adj(m, ip_len - m->m_pkthdr.len); } /* * Is packet from or to 127/8? */ if ((ntohl(ip->ip_dst.s_addr) >> IN_CLASSA_NSHIFT) == IN_LOOPBACKNET || (ntohl(ip->ip_src.s_addr) >> IN_CLASSA_NSHIFT) == IN_LOOPBACKNET) { IPSTAT_INC(ips_badaddr); goto drop; } #ifdef ALTQ /* * Is packet dropped by traffic conditioner? */ if (altq_input != NULL && (*altq_input)(m, AF_INET) == 0) goto drop; #endif /* * Step 2: fallback conditions to normal ip_input path processing */ /* * Only IP packets without options */ if (ip->ip_hl != (sizeof(struct ip) >> 2)) { - if (ip_doopts == 1) + if (V_ip_doopts == 1) return m; - else if (ip_doopts == 2) { + else if (V_ip_doopts == 2) { icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_FILTER_PROHIB, 0, 0); return NULL; /* mbuf already free'd */ } /* else ignore IP options and continue */ } /* * Only unicast IP, not from loopback, no L2 or IP broadcast, * no multicast, no INADDR_ANY * * XXX: Probably some of these checks could be direct drop * conditions. However it is not clear whether there are some * hacks or obscure behaviours which make it neccessary to * let ip_input handle it. We play safe here and let ip_input * deal with it until it is proven that we can directly drop it. */ if ((m->m_flags & (M_BCAST|M_MCAST)) || (m->m_pkthdr.rcvif->if_flags & IFF_LOOPBACK) || ntohl(ip->ip_src.s_addr) == (u_long)INADDR_BROADCAST || ntohl(ip->ip_dst.s_addr) == (u_long)INADDR_BROADCAST || IN_MULTICAST(ntohl(ip->ip_src.s_addr)) || IN_MULTICAST(ntohl(ip->ip_dst.s_addr)) || IN_LINKLOCAL(ntohl(ip->ip_src.s_addr)) || IN_LINKLOCAL(ntohl(ip->ip_dst.s_addr)) || ip->ip_src.s_addr == INADDR_ANY || ip->ip_dst.s_addr == INADDR_ANY ) return m; /* * Is it for a local address on this host? */ if (in_localip(ip->ip_dst)) return m; IPSTAT_INC(ips_total); /* * Step 3: incoming packet firewall processing */ odest.s_addr = dest.s_addr = ip->ip_dst.s_addr; /* * Run through list of ipfilter hooks for input packets */ if (!PFIL_HOOKED(&V_inet_pfil_hook)) goto passin; if (pfil_run_hooks( &V_inet_pfil_hook, &m, m->m_pkthdr.rcvif, PFIL_IN, NULL) || m == NULL) goto drop; M_ASSERTVALID(m); M_ASSERTPKTHDR(m); ip = mtod(m, struct ip *); /* m may have changed by pfil hook */ dest.s_addr = ip->ip_dst.s_addr; /* * Destination address changed? */ if (odest.s_addr != dest.s_addr) { /* * Is it now for a local address on this host? */ if (in_localip(dest)) goto forwardlocal; /* * Go on with new destination address */ } if (m->m_flags & M_FASTFWD_OURS) { /* * ipfw changed it for a local address on this host. */ goto forwardlocal; } passin: /* * Step 4: decrement TTL and look up route */ /* * Check TTL */ #ifdef IPSTEALTH if (!V_ipstealth) { #endif if (ip->ip_ttl <= IPTTLDEC) { icmp_error(m, ICMP_TIMXCEED, ICMP_TIMXCEED_INTRANS, 0, 0); return NULL; /* mbuf already free'd */ } /* * Decrement the TTL and incrementally change the IP header checksum. * Don't bother doing this with hw checksum offloading, it's faster * doing it right here. */ ip->ip_ttl -= IPTTLDEC; if (ip->ip_sum >= (u_int16_t) ~htons(IPTTLDEC << 8)) ip->ip_sum -= ~htons(IPTTLDEC << 8); else ip->ip_sum += htons(IPTTLDEC << 8); #ifdef IPSTEALTH } #endif /* * Find route to destination. */ if ((dst = ip_findroute(&ro, dest, m)) == NULL) return NULL; /* icmp unreach already sent */ ifp = ro.ro_rt->rt_ifp; /* * Immediately drop blackholed traffic, and directed broadcasts * for either the all-ones or all-zero subnet addresses on * locally attached networks. */ if ((ro.ro_rt->rt_flags & (RTF_BLACKHOLE|RTF_BROADCAST)) != 0) goto drop; /* * Step 5: outgoing firewall packet processing */ /* * Run through list of hooks for output packets. */ if (!PFIL_HOOKED(&V_inet_pfil_hook)) goto passout; if (pfil_run_hooks(&V_inet_pfil_hook, &m, ifp, PFIL_OUT, NULL) || m == NULL) { goto drop; } M_ASSERTVALID(m); M_ASSERTPKTHDR(m); ip = mtod(m, struct ip *); dest.s_addr = ip->ip_dst.s_addr; /* * Destination address changed? */ if (m->m_flags & M_IP_NEXTHOP) fwd_tag = m_tag_find(m, PACKET_TAG_IPFORWARD, NULL); if (odest.s_addr != dest.s_addr || fwd_tag != NULL) { /* * Is it now for a local address on this host? */ if (m->m_flags & M_FASTFWD_OURS || in_localip(dest)) { forwardlocal: /* * Return packet for processing by ip_input(). */ m->m_flags |= M_FASTFWD_OURS; if (ro.ro_rt) RTFREE(ro.ro_rt); return m; } /* * Redo route lookup with new destination address */ if (fwd_tag) { dest.s_addr = ((struct sockaddr_in *) (fwd_tag + 1))->sin_addr.s_addr; m_tag_delete(m, fwd_tag); m->m_flags &= ~M_IP_NEXTHOP; } RTFREE(ro.ro_rt); if ((dst = ip_findroute(&ro, dest, m)) == NULL) return NULL; /* icmp unreach already sent */ ifp = ro.ro_rt->rt_ifp; } passout: /* * Step 6: send off the packet */ ip_len = ntohs(ip->ip_len); ip_off = ntohs(ip->ip_off); /* * Check if route is dampned (when ARP is unable to resolve) */ if ((ro.ro_rt->rt_flags & RTF_REJECT) && (ro.ro_rt->rt_expire == 0 || time_uptime < ro.ro_rt->rt_expire)) { icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_HOST, 0, 0); goto consumed; } #ifndef ALTQ /* * Check if there is enough space in the interface queue */ if ((ifp->if_snd.ifq_len + ip_len / ifp->if_mtu + 1) >= ifp->if_snd.ifq_maxlen) { IPSTAT_INC(ips_odropped); /* would send source quench here but that is depreciated */ goto drop; } #endif /* * Check if media link state of interface is not down */ if (ifp->if_link_state == LINK_STATE_DOWN) { icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_HOST, 0, 0); goto consumed; } /* * Check if packet fits MTU or if hardware will fragment for us */ if (ro.ro_rt->rt_mtu) mtu = min(ro.ro_rt->rt_mtu, ifp->if_mtu); else mtu = ifp->if_mtu; if (ip_len <= mtu || (ifp->if_hwassist & CSUM_FRAGMENT && (ip_off & IP_DF) == 0)) { /* * Avoid confusing lower layers. */ m_clrprotoflags(m); /* * Send off the packet via outgoing interface */ IP_PROBE(send, NULL, NULL, ip, ifp, ip, NULL); error = (*ifp->if_output)(ifp, m, (struct sockaddr *)dst, &ro); } else { /* * Handle EMSGSIZE with icmp reply needfrag for TCP MTU discovery */ if (ip_off & IP_DF) { IPSTAT_INC(ips_cantfrag); icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_NEEDFRAG, 0, mtu); goto consumed; } else { /* * We have to fragment the packet */ m->m_pkthdr.csum_flags |= CSUM_IP; if (ip_fragment(ip, &m, mtu, ifp->if_hwassist)) goto drop; KASSERT(m != NULL, ("null mbuf and no error")); /* * Send off the fragments via outgoing interface */ error = 0; do { m0 = m->m_nextpkt; m->m_nextpkt = NULL; /* * Avoid confusing lower layers. */ m_clrprotoflags(m); IP_PROBE(send, NULL, NULL, ip, ifp, ip, NULL); error = (*ifp->if_output)(ifp, m, (struct sockaddr *)dst, &ro); if (error) break; } while ((m = m0) != NULL); if (error) { /* Reclaim remaining fragments */ for (m = m0; m; m = m0) { m0 = m->m_nextpkt; m_freem(m); } } else IPSTAT_INC(ips_fragmented); } } if (error != 0) IPSTAT_INC(ips_odropped); else { counter_u64_add(ro.ro_rt->rt_pksent, 1); IPSTAT_INC(ips_forward); IPSTAT_INC(ips_fastforward); } consumed: RTFREE(ro.ro_rt); return NULL; drop: if (m) m_freem(m); if (ro.ro_rt) RTFREE(ro.ro_rt); return NULL; } Index: stable/10/sys/netinet/ip_options.c =================================================================== --- stable/10/sys/netinet/ip_options.c (revision 272867) +++ stable/10/sys/netinet/ip_options.c (revision 272868) @@ -1,737 +1,740 @@ /* * Copyright (c) 1982, 1986, 1988, 1993 * The Regents of the University of California. * Copyright (c) 2005 Andre Oppermann, Internet Business Solutions AG. * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 4. Neither the name of the University nor the names of its contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. */ #include __FBSDID("$FreeBSD$"); #include "opt_ipstealth.h" #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include -static int ip_dosourceroute = 0; -SYSCTL_INT(_net_inet_ip, IPCTL_SOURCEROUTE, sourceroute, CTLFLAG_RW, - &ip_dosourceroute, 0, "Enable forwarding source routed IP packets"); +static VNET_DEFINE(int, ip_dosourceroute); +SYSCTL_VNET_INT(_net_inet_ip, IPCTL_SOURCEROUTE, sourceroute, CTLFLAG_RW, + &VNET_NAME(ip_dosourceroute), 0, + "Enable forwarding source routed IP packets"); +#define V_ip_dosourceroute VNET(ip_dosourceroute) -static int ip_acceptsourceroute = 0; -SYSCTL_INT(_net_inet_ip, IPCTL_ACCEPTSOURCEROUTE, accept_sourceroute, - CTLFLAG_RW, &ip_acceptsourceroute, 0, +static VNET_DEFINE(int, ip_acceptsourceroute); +SYSCTL_VNET_INT(_net_inet_ip, IPCTL_ACCEPTSOURCEROUTE, accept_sourceroute, + CTLFLAG_RW, &VNET_NAME(ip_acceptsourceroute), 0, "Enable accepting source routed IP packets"); +#define V_ip_acceptsourceroute VNET(ip_acceptsourceroute) -int ip_doopts = 1; /* 0 = ignore, 1 = process, 2 = reject */ -SYSCTL_INT(_net_inet_ip, OID_AUTO, process_options, CTLFLAG_RW, - &ip_doopts, 0, "Enable IP options processing ([LS]SRR, RR, TS)"); +VNET_DEFINE(int, ip_doopts) = 1; /* 0 = ignore, 1 = process, 2 = reject */ +SYSCTL_VNET_INT(_net_inet_ip, OID_AUTO, process_options, CTLFLAG_RW, + &VNET_NAME(ip_doopts), 0, "Enable IP options processing ([LS]SRR, RR, TS)"); static void save_rte(struct mbuf *m, u_char *, struct in_addr); /* * Do option processing on a datagram, possibly discarding it if bad options * are encountered, or forwarding it if source-routed. * * The pass argument is used when operating in the IPSTEALTH mode to tell * what options to process: [LS]SRR (pass 0) or the others (pass 1). The * reason for as many as two passes is that when doing IPSTEALTH, non-routing * options should be processed only if the packet is for us. * * Returns 1 if packet has been forwarded/freed, 0 if the packet should be * processed further. */ int ip_dooptions(struct mbuf *m, int pass) { struct ip *ip = mtod(m, struct ip *); u_char *cp; struct in_ifaddr *ia; int opt, optlen, cnt, off, code, type = ICMP_PARAMPROB, forward = 0; struct in_addr *sin, dst; uint32_t ntime; struct sockaddr_in ipaddr = { sizeof(ipaddr), AF_INET }; /* Ignore or reject packets with IP options. */ - if (ip_doopts == 0) + if (V_ip_doopts == 0) return 0; - else if (ip_doopts == 2) { + else if (V_ip_doopts == 2) { type = ICMP_UNREACH; code = ICMP_UNREACH_FILTER_PROHIB; goto bad; } dst = ip->ip_dst; cp = (u_char *)(ip + 1); cnt = (ip->ip_hl << 2) - sizeof (struct ip); for (; cnt > 0; cnt -= optlen, cp += optlen) { opt = cp[IPOPT_OPTVAL]; if (opt == IPOPT_EOL) break; if (opt == IPOPT_NOP) optlen = 1; else { if (cnt < IPOPT_OLEN + sizeof(*cp)) { code = &cp[IPOPT_OLEN] - (u_char *)ip; goto bad; } optlen = cp[IPOPT_OLEN]; if (optlen < IPOPT_OLEN + sizeof(*cp) || optlen > cnt) { code = &cp[IPOPT_OLEN] - (u_char *)ip; goto bad; } } switch (opt) { default: break; /* * Source routing with record. Find interface with current * destination address. If none on this machine then drop if * strictly routed, or do nothing if loosely routed. Record * interface address and bring up next address component. If * strictly routed make sure next address is on directly * accessible net. */ case IPOPT_LSRR: case IPOPT_SSRR: #ifdef IPSTEALTH if (V_ipstealth && pass > 0) break; #endif if (optlen < IPOPT_OFFSET + sizeof(*cp)) { code = &cp[IPOPT_OLEN] - (u_char *)ip; goto bad; } if ((off = cp[IPOPT_OFFSET]) < IPOPT_MINOFF) { code = &cp[IPOPT_OFFSET] - (u_char *)ip; goto bad; } ipaddr.sin_addr = ip->ip_dst; if (ifa_ifwithaddr_check((struct sockaddr *)&ipaddr) == 0) { if (opt == IPOPT_SSRR) { type = ICMP_UNREACH; code = ICMP_UNREACH_SRCFAIL; goto bad; } - if (!ip_dosourceroute) + if (!V_ip_dosourceroute) goto nosourcerouting; /* * Loose routing, and not at next destination * yet; nothing to do except forward. */ break; } off--; /* 0 origin */ if (off > optlen - (int)sizeof(struct in_addr)) { /* * End of source route. Should be for us. */ - if (!ip_acceptsourceroute) + if (!V_ip_acceptsourceroute) goto nosourcerouting; save_rte(m, cp, ip->ip_src); break; } #ifdef IPSTEALTH if (V_ipstealth) goto dropit; #endif - if (!ip_dosourceroute) { + if (!V_ip_dosourceroute) { if (V_ipforwarding) { char buf[16]; /* aaa.bbb.ccc.ddd\0 */ /* * Acting as a router, so generate * ICMP */ nosourcerouting: strcpy(buf, inet_ntoa(ip->ip_dst)); log(LOG_WARNING, "attempted source route from %s to %s\n", inet_ntoa(ip->ip_src), buf); type = ICMP_UNREACH; code = ICMP_UNREACH_SRCFAIL; goto bad; } else { /* * Not acting as a router, so * silently drop. */ #ifdef IPSTEALTH dropit: #endif IPSTAT_INC(ips_cantforward); m_freem(m); return (1); } } /* * locate outgoing interface */ (void)memcpy(&ipaddr.sin_addr, cp + off, sizeof(ipaddr.sin_addr)); if (opt == IPOPT_SSRR) { #define INA struct in_ifaddr * #define SA struct sockaddr * if ((ia = (INA)ifa_ifwithdstaddr((SA)&ipaddr)) == NULL) ia = (INA)ifa_ifwithnet((SA)&ipaddr, 0); } else /* XXX MRT 0 for routing */ ia = ip_rtaddr(ipaddr.sin_addr, M_GETFIB(m)); if (ia == NULL) { type = ICMP_UNREACH; code = ICMP_UNREACH_SRCFAIL; goto bad; } ip->ip_dst = ipaddr.sin_addr; (void)memcpy(cp + off, &(IA_SIN(ia)->sin_addr), sizeof(struct in_addr)); ifa_free(&ia->ia_ifa); cp[IPOPT_OFFSET] += sizeof(struct in_addr); /* * Let ip_intr's mcast routing check handle mcast pkts */ forward = !IN_MULTICAST(ntohl(ip->ip_dst.s_addr)); break; case IPOPT_RR: #ifdef IPSTEALTH if (V_ipstealth && pass == 0) break; #endif if (optlen < IPOPT_OFFSET + sizeof(*cp)) { code = &cp[IPOPT_OFFSET] - (u_char *)ip; goto bad; } if ((off = cp[IPOPT_OFFSET]) < IPOPT_MINOFF) { code = &cp[IPOPT_OFFSET] - (u_char *)ip; goto bad; } /* * If no space remains, ignore. */ off--; /* 0 origin */ if (off > optlen - (int)sizeof(struct in_addr)) break; (void)memcpy(&ipaddr.sin_addr, &ip->ip_dst, sizeof(ipaddr.sin_addr)); /* * Locate outgoing interface; if we're the * destination, use the incoming interface (should be * same). */ if ((ia = (INA)ifa_ifwithaddr((SA)&ipaddr)) == NULL && (ia = ip_rtaddr(ipaddr.sin_addr, M_GETFIB(m))) == NULL) { type = ICMP_UNREACH; code = ICMP_UNREACH_HOST; goto bad; } (void)memcpy(cp + off, &(IA_SIN(ia)->sin_addr), sizeof(struct in_addr)); ifa_free(&ia->ia_ifa); cp[IPOPT_OFFSET] += sizeof(struct in_addr); break; case IPOPT_TS: #ifdef IPSTEALTH if (V_ipstealth && pass == 0) break; #endif code = cp - (u_char *)ip; if (optlen < 4 || optlen > 40) { code = &cp[IPOPT_OLEN] - (u_char *)ip; goto bad; } if ((off = cp[IPOPT_OFFSET]) < 5) { code = &cp[IPOPT_OLEN] - (u_char *)ip; goto bad; } if (off > optlen - (int)sizeof(int32_t)) { cp[IPOPT_OFFSET + 1] += (1 << 4); if ((cp[IPOPT_OFFSET + 1] & 0xf0) == 0) { code = &cp[IPOPT_OFFSET] - (u_char *)ip; goto bad; } break; } off--; /* 0 origin */ sin = (struct in_addr *)(cp + off); switch (cp[IPOPT_OFFSET + 1] & 0x0f) { case IPOPT_TS_TSONLY: break; case IPOPT_TS_TSANDADDR: if (off + sizeof(uint32_t) + sizeof(struct in_addr) > optlen) { code = &cp[IPOPT_OFFSET] - (u_char *)ip; goto bad; } ipaddr.sin_addr = dst; ia = (INA)ifaof_ifpforaddr((SA)&ipaddr, m->m_pkthdr.rcvif); if (ia == NULL) continue; (void)memcpy(sin, &IA_SIN(ia)->sin_addr, sizeof(struct in_addr)); ifa_free(&ia->ia_ifa); cp[IPOPT_OFFSET] += sizeof(struct in_addr); off += sizeof(struct in_addr); break; case IPOPT_TS_PRESPEC: if (off + sizeof(uint32_t) + sizeof(struct in_addr) > optlen) { code = &cp[IPOPT_OFFSET] - (u_char *)ip; goto bad; } (void)memcpy(&ipaddr.sin_addr, sin, sizeof(struct in_addr)); if (ifa_ifwithaddr_check((SA)&ipaddr) == 0) continue; cp[IPOPT_OFFSET] += sizeof(struct in_addr); off += sizeof(struct in_addr); break; default: code = &cp[IPOPT_OFFSET + 1] - (u_char *)ip; goto bad; } ntime = iptime(); (void)memcpy(cp + off, &ntime, sizeof(uint32_t)); cp[IPOPT_OFFSET] += sizeof(uint32_t); } } if (forward && V_ipforwarding) { ip_forward(m, 1); return (1); } return (0); bad: icmp_error(m, type, code, 0, 0); IPSTAT_INC(ips_badoptions); return (1); } /* * Save incoming source route for use in replies, to be picked up later by * ip_srcroute if the receiver is interested. */ static void save_rte(struct mbuf *m, u_char *option, struct in_addr dst) { unsigned olen; struct ipopt_tag *opts; opts = (struct ipopt_tag *)m_tag_get(PACKET_TAG_IPOPTIONS, sizeof(struct ipopt_tag), M_NOWAIT); if (opts == NULL) return; olen = option[IPOPT_OLEN]; if (olen > sizeof(opts->ip_srcrt) - (1 + sizeof(dst))) { m_tag_free((struct m_tag *)opts); return; } bcopy(option, opts->ip_srcrt.srcopt, olen); opts->ip_nhops = (olen - IPOPT_OFFSET - 1) / sizeof(struct in_addr); opts->ip_srcrt.dst = dst; m_tag_prepend(m, (struct m_tag *)opts); } /* * Retrieve incoming source route for use in replies, in the same form used * by setsockopt. The first hop is placed before the options, will be * removed later. */ struct mbuf * ip_srcroute(struct mbuf *m0) { struct in_addr *p, *q; struct mbuf *m; struct ipopt_tag *opts; opts = (struct ipopt_tag *)m_tag_find(m0, PACKET_TAG_IPOPTIONS, NULL); if (opts == NULL) return (NULL); if (opts->ip_nhops == 0) return (NULL); m = m_get(M_NOWAIT, MT_DATA); if (m == NULL) return (NULL); #define OPTSIZ (sizeof(opts->ip_srcrt.nop) + sizeof(opts->ip_srcrt.srcopt)) /* length is (nhops+1)*sizeof(addr) + sizeof(nop + srcrt header) */ m->m_len = opts->ip_nhops * sizeof(struct in_addr) + sizeof(struct in_addr) + OPTSIZ; /* * First, save first hop for return route. */ p = &(opts->ip_srcrt.route[opts->ip_nhops - 1]); *(mtod(m, struct in_addr *)) = *p--; /* * Copy option fields and padding (nop) to mbuf. */ opts->ip_srcrt.nop = IPOPT_NOP; opts->ip_srcrt.srcopt[IPOPT_OFFSET] = IPOPT_MINOFF; (void)memcpy(mtod(m, caddr_t) + sizeof(struct in_addr), &(opts->ip_srcrt.nop), OPTSIZ); q = (struct in_addr *)(mtod(m, caddr_t) + sizeof(struct in_addr) + OPTSIZ); #undef OPTSIZ /* * Record return path as an IP source route, reversing the path * (pointers are now aligned). */ while (p >= opts->ip_srcrt.route) { *q++ = *p--; } /* * Last hop goes to final destination. */ *q = opts->ip_srcrt.dst; m_tag_delete(m0, (struct m_tag *)opts); return (m); } /* * Strip out IP options, at higher level protocol in the kernel. */ void ip_stripoptions(struct mbuf *m) { struct ip *ip = mtod(m, struct ip *); int olen; olen = (ip->ip_hl << 2) - sizeof(struct ip); m->m_len -= olen; if (m->m_flags & M_PKTHDR) m->m_pkthdr.len -= olen; ip->ip_len = htons(ntohs(ip->ip_len) - olen); ip->ip_hl = sizeof(struct ip) >> 2; bcopy((char *)ip + sizeof(struct ip) + olen, (ip + 1), (size_t )(m->m_len - sizeof(struct ip))); } /* * Insert IP options into preformed packet. Adjust IP destination as * required for IP source routing, as indicated by a non-zero in_addr at the * start of the options. * * XXX This routine assumes that the packet has no options in place. */ struct mbuf * ip_insertoptions(struct mbuf *m, struct mbuf *opt, int *phlen) { struct ipoption *p = mtod(opt, struct ipoption *); struct mbuf *n; struct ip *ip = mtod(m, struct ip *); unsigned optlen; optlen = opt->m_len - sizeof(p->ipopt_dst); if (optlen + ntohs(ip->ip_len) > IP_MAXPACKET) { *phlen = 0; return (m); /* XXX should fail */ } if (p->ipopt_dst.s_addr) ip->ip_dst = p->ipopt_dst; if (m->m_flags & M_EXT || m->m_data - optlen < m->m_pktdat) { n = m_gethdr(M_NOWAIT, MT_DATA); if (n == NULL) { *phlen = 0; return (m); } m_move_pkthdr(n, m); n->m_pkthdr.rcvif = NULL; n->m_pkthdr.len += optlen; m->m_len -= sizeof(struct ip); m->m_data += sizeof(struct ip); n->m_next = m; m = n; m->m_len = optlen + sizeof(struct ip); m->m_data += max_linkhdr; bcopy(ip, mtod(m, void *), sizeof(struct ip)); } else { m->m_data -= optlen; m->m_len += optlen; m->m_pkthdr.len += optlen; bcopy(ip, mtod(m, void *), sizeof(struct ip)); } ip = mtod(m, struct ip *); bcopy(p->ipopt_list, ip + 1, optlen); *phlen = sizeof(struct ip) + optlen; ip->ip_v = IPVERSION; ip->ip_hl = *phlen >> 2; ip->ip_len = htons(ntohs(ip->ip_len) + optlen); return (m); } /* * Copy options from ip to jp, omitting those not copied during * fragmentation. */ int ip_optcopy(struct ip *ip, struct ip *jp) { u_char *cp, *dp; int opt, optlen, cnt; cp = (u_char *)(ip + 1); dp = (u_char *)(jp + 1); cnt = (ip->ip_hl << 2) - sizeof (struct ip); for (; cnt > 0; cnt -= optlen, cp += optlen) { opt = cp[0]; if (opt == IPOPT_EOL) break; if (opt == IPOPT_NOP) { /* Preserve for IP mcast tunnel's LSRR alignment. */ *dp++ = IPOPT_NOP; optlen = 1; continue; } KASSERT(cnt >= IPOPT_OLEN + sizeof(*cp), ("ip_optcopy: malformed ipv4 option")); optlen = cp[IPOPT_OLEN]; KASSERT(optlen >= IPOPT_OLEN + sizeof(*cp) && optlen <= cnt, ("ip_optcopy: malformed ipv4 option")); /* Bogus lengths should have been caught by ip_dooptions. */ if (optlen > cnt) optlen = cnt; if (IPOPT_COPIED(opt)) { bcopy(cp, dp, optlen); dp += optlen; } } for (optlen = dp - (u_char *)(jp+1); optlen & 0x3; optlen++) *dp++ = IPOPT_EOL; return (optlen); } /* * Set up IP options in pcb for insertion in output packets. Store in mbuf * with pointer in pcbopt, adding pseudo-option with destination address if * source routed. */ int ip_pcbopts(struct inpcb *inp, int optname, struct mbuf *m) { int cnt, optlen; u_char *cp; struct mbuf **pcbopt; u_char opt; INP_WLOCK_ASSERT(inp); pcbopt = &inp->inp_options; /* turn off any old options */ if (*pcbopt) (void)m_free(*pcbopt); *pcbopt = 0; if (m == NULL || m->m_len == 0) { /* * Only turning off any previous options. */ if (m != NULL) (void)m_free(m); return (0); } if (m->m_len % sizeof(int32_t)) goto bad; /* * IP first-hop destination address will be stored before actual * options; move other options back and clear it when none present. */ if (m->m_data + m->m_len + sizeof(struct in_addr) >= &m->m_dat[MLEN]) goto bad; cnt = m->m_len; m->m_len += sizeof(struct in_addr); cp = mtod(m, u_char *) + sizeof(struct in_addr); bcopy(mtod(m, void *), cp, (unsigned)cnt); bzero(mtod(m, void *), sizeof(struct in_addr)); for (; cnt > 0; cnt -= optlen, cp += optlen) { opt = cp[IPOPT_OPTVAL]; if (opt == IPOPT_EOL) break; if (opt == IPOPT_NOP) optlen = 1; else { if (cnt < IPOPT_OLEN + sizeof(*cp)) goto bad; optlen = cp[IPOPT_OLEN]; if (optlen < IPOPT_OLEN + sizeof(*cp) || optlen > cnt) goto bad; } switch (opt) { default: break; case IPOPT_LSRR: case IPOPT_SSRR: /* * User process specifies route as: * * ->A->B->C->D * * D must be our final destination (but we can't * check that since we may not have connected yet). * A is first hop destination, which doesn't appear * in actual IP option, but is stored before the * options. */ /* XXX-BZ PRIV_NETINET_SETHDROPTS? */ if (optlen < IPOPT_MINOFF - 1 + sizeof(struct in_addr)) goto bad; m->m_len -= sizeof(struct in_addr); cnt -= sizeof(struct in_addr); optlen -= sizeof(struct in_addr); cp[IPOPT_OLEN] = optlen; /* * Move first hop before start of options. */ bcopy((caddr_t)&cp[IPOPT_OFFSET+1], mtod(m, caddr_t), sizeof(struct in_addr)); /* * Then copy rest of options back * to close up the deleted entry. */ bcopy((&cp[IPOPT_OFFSET+1] + sizeof(struct in_addr)), &cp[IPOPT_OFFSET+1], (unsigned)cnt - (IPOPT_MINOFF - 1)); break; } } if (m->m_len > MAX_IPOPTLEN + sizeof(struct in_addr)) goto bad; *pcbopt = m; return (0); bad: (void)m_free(m); return (EINVAL); } /* * Check for the presence of the IP Router Alert option [RFC2113] * in the header of an IPv4 datagram. * * This call is not intended for use from the forwarding path; it is here * so that protocol domains may check for the presence of the option. * Given how FreeBSD's IPv4 stack is currently structured, the Router Alert * option does not have much relevance to the implementation, though this * may change in future. * Router alert options SHOULD be passed if running in IPSTEALTH mode and * we are not the endpoint. * Length checks on individual options should already have been peformed * by ip_dooptions() therefore they are folded under INVARIANTS here. * * Return zero if not present or options are invalid, non-zero if present. */ int ip_checkrouteralert(struct mbuf *m) { struct ip *ip = mtod(m, struct ip *); u_char *cp; int opt, optlen, cnt, found_ra; found_ra = 0; cp = (u_char *)(ip + 1); cnt = (ip->ip_hl << 2) - sizeof (struct ip); for (; cnt > 0; cnt -= optlen, cp += optlen) { opt = cp[IPOPT_OPTVAL]; if (opt == IPOPT_EOL) break; if (opt == IPOPT_NOP) optlen = 1; else { #ifdef INVARIANTS if (cnt < IPOPT_OLEN + sizeof(*cp)) break; #endif optlen = cp[IPOPT_OLEN]; #ifdef INVARIANTS if (optlen < IPOPT_OLEN + sizeof(*cp) || optlen > cnt) break; #endif } switch (opt) { case IPOPT_RA: #ifdef INVARIANTS if (optlen != IPOPT_OFFSET + sizeof(uint16_t) || (*((uint16_t *)&cp[IPOPT_OFFSET]) != 0)) break; else #endif found_ra = 1; break; default: break; } } return (found_ra); } Index: stable/10/sys/netinet/ip_options.h =================================================================== --- stable/10/sys/netinet/ip_options.h (revision 272867) +++ stable/10/sys/netinet/ip_options.h (revision 272868) @@ -1,60 +1,61 @@ /* * Copyright (c) 1982, 1986, 1993 * The Regents of the University of California. * Copyright (c) 2005 Andre Oppermann, Internet Business Solutions AG. * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 4. Neither the name of the University nor the names of its contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. * * $FreeBSD$ */ #ifndef _NETINET_IP_OPTIONS_H_ #define _NETINET_IP_OPTIONS_H_ struct ipoptrt { struct in_addr dst; /* final destination */ char nop; /* one NOP to align */ char srcopt[IPOPT_OFFSET + 1]; /* OPTVAL, OLEN and OFFSET */ struct in_addr route[MAX_IPOPTLEN/sizeof(struct in_addr)]; }; struct ipopt_tag { struct m_tag tag; /* m_tag */ int ip_nhops; struct ipoptrt ip_srcrt; }; -extern int ip_doopts; /* process or ignore IP options */ +VNET_DECLARE(int, ip_doopts); /* process or ignore IP options */ +#define V_ip_doopts VNET(ip_doopts) int ip_checkrouteralert(struct mbuf *); int ip_dooptions(struct mbuf *, int); struct mbuf *ip_insertoptions(struct mbuf *, struct mbuf *, int *); int ip_optcopy(struct ip *, struct ip *); int ip_pcbopts(struct inpcb *, int, struct mbuf *); void ip_stripoptions(struct mbuf *); struct mbuf *ip_srcroute(struct mbuf *); #endif /* !_NETINET_IP_OPTIONS_H_ */ Index: stable/10 =================================================================== --- stable/10 (revision 272867) +++ stable/10 (revision 272868) Property changes on: stable/10 ___________________________________________________________________ Modified: svn:mergeinfo ## -0,0 +0,1 ## Merged /head:r271545,271610