#! /bin/sh # Spectre & Meltdown checker # # Check for the latest version at: # https://github.com/speed47/spectre-meltdown-checker # git clone https://github.com/speed47/spectre-meltdown-checker.git # or wget https://meltdown.ovh -O spectre-meltdown-checker.sh # or curl -L https://meltdown.ovh -o spectre-meltdown-checker.sh # # Stephane Lesimple # VERSION='0.36+' trap 'exit_cleanup' EXIT trap '_warn "interrupted, cleaning up..."; exit_cleanup; exit 1' INT exit_cleanup() { # cleanup the temp decompressed config & kernel image [ -n "$dumped_config" ] && [ -f "$dumped_config" ] && rm -f "$dumped_config" [ -n "$kerneltmp" ] && [ -f "$kerneltmp" ] && rm -f "$kerneltmp" [ -n "$kerneltmp2" ] && [ -f "$kerneltmp2" ] && rm -f "$kerneltmp2" [ "$mounted_debugfs" = 1 ] && umount /sys/kernel/debug 2>/dev/null [ "$mounted_procfs" = 1 ] && umount "$procfs" 2>/dev/null [ "$insmod_cpuid" = 1 ] && rmmod cpuid 2>/dev/null [ "$insmod_msr" = 1 ] && rmmod msr 2>/dev/null [ "$kldload_cpuctl" = 1 ] && kldunload cpuctl 2>/dev/null } show_usage() { # shellcheck disable=SC2086 cat <] [--config ] [--map ] Modes: Two modes are available. First mode is the "live" mode (default), it does its best to find information about the currently running kernel. To run under this mode, just start the script without any option (you can also use --live explicitly) Second mode is the "offline" mode, where you can inspect a non-running kernel. You'll need to specify the location of the kernel file, config and System.map files: --kernel kernel_file specify a (possibly compressed) Linux or BSD kernel file --config kernel_config specify a kernel config file (Linux only) --map kernel_map_file specify a kernel System.map file (Linux only) Options: --no-color don't use color codes --verbose, -v increase verbosity level, possibly several times --no-explain don't produce a human-readable explanation of actions to take to mitigate a vulnerability --no-sysfs don't use the /sys interface even if present [Linux] --sysfs-only only use the /sys interface, don't run our own checks [Linux] --coreos special mode for CoreOS (use an ephemeral toolbox to inspect kernel) [Linux] --arch-prefix PREFIX specify a prefix for cross-inspecting a kernel of a different arch, for example "aarch64-linux-gnu-", so that invoked tools will be prefixed with this (i.e. aarch64-linux-gnu-objdump) --batch text produce machine readable output, this is the default if --batch is specified alone --batch json produce JSON output formatted for Puppet, Ansible, Chef... --batch nrpe produce machine readable output formatted for NRPE --batch prometheus produce output for consumption by prometheus-node-exporter --variant [1,2,3] specify which variant you'd like to check, by default all variants are checked, can be specified multiple times (e.g. --variant 2 --variant 3) --hw-only only check for CPU information, don't check for any variant --no-hw skip CPU information and checks, if you're inspecting a kernel not to be run on this host Return codes: 0 (not vulnerable), 2 (vulnerable), 3 (unknown), 255 (error) IMPORTANT: A false sense of security is worse than no security at all. Please use the --disclaimer option to understand exactly what this script does. EOF } show_disclaimer() { cat </dev/null 2>&1; then echo_cmd=$(which printf) echo_cmd_type=printf elif which echo >/dev/null 2>&1; then echo_cmd=$(which echo) else # which command is broken? [ -x /bin/echo ] && echo_cmd=/bin/echo # for Android [ -x /system/bin/echo ] && echo_cmd=/system/bin/echo fi # still empty ? fallback to builtin [ -z "$echo_cmd" ] && echo_cmd=echo __echo() { opt="$1" shift _msg="$*" if [ "$opt_no_color" = 1 ] ; then # strip ANSI color codes # some sed versions (i.e. toybox) can't seem to handle # \033 aka \x1B correctly, so do it for them. if [ "$echo_cmd_type" = printf ]; then _interpret_chars='' else _interpret_chars='-e' fi _ctrlchar=$($echo_cmd $_interpret_chars "\033") _msg=$($echo_cmd $_interpret_chars "$_msg" | sed -r "s/$_ctrlchar\[([0-9][0-9]?(;[0-9][0-9]?)?)?m//g") fi if [ "$echo_cmd_type" = printf ]; then if [ "$opt" = "-n" ]; then $echo_cmd "$_msg" else $echo_cmd "$_msg\n" fi else # shellcheck disable=SC2086 $echo_cmd $opt -e "$_msg" fi } _echo() { if [ "$opt_verbose" -ge "$1" ]; then shift __echo '' "$*" fi } _echo_nol() { if [ "$opt_verbose" -ge "$1" ]; then shift __echo -n "$*" fi } _warn() { _echo 0 "\033[31m$*\033[0m" >&2 } _info() { _echo 1 "$*" } _info_nol() { _echo_nol 1 "$*" } _verbose() { _echo 2 "$*" } _verbose_nol() { _echo_nol 2 "$*" } _debug() { _echo 3 "\033[34m(debug) $*\033[0m" } explain() { if [ "$opt_no_explain" != 1 ] ; then _info '' _info "> \033[41m\033[30mWhat you need to do:\033[0m $*" fi } is_cpu_vulnerable_cached=0 _is_cpu_vulnerable_cached() { # shellcheck disable=SC2086 [ "$1" = 1 ] && return $variant1 # shellcheck disable=SC2086 [ "$1" = 2 ] && return $variant2 # shellcheck disable=SC2086 [ "$1" = 3 ] && return $variant3 echo "$0: error: invalid variant '$1' passed to is_cpu_vulnerable()" >&2 exit 255 } is_cpu_vulnerable() { # param: 1, 2 or 3 (variant) # returns 0 if vulnerable, 1 if not vulnerable # (note that in shell, a return of 0 is success) # by default, everything is vulnerable, we work in a "whitelist" logic here. # usage: is_cpu_vulnerable 2 && do something if vulnerable if [ "$is_cpu_vulnerable_cached" = 1 ]; then _is_cpu_vulnerable_cached "$1" return $? fi variant1='' variant2='' variant3='' if is_cpu_specex_free; then variant1=immune variant2=immune variant3=immune elif [ "$cpu_vendor" = GenuineIntel ]; then # Intel # https://github.com/crozone/SpectrePoC/issues/1 ^F E5200 => spectre 2 not vulnerable # https://github.com/paboldin/meltdown-exploit/issues/19 ^F E5200 => meltdown vulnerable # model name : Pentium(R) Dual-Core CPU E5200 @ 2.50GHz if grep -qE '^model name.+ Pentium\(R\) Dual-Core[[:space:]]+CPU[[:space:]]+E[0-9]{4}K? ' "$procfs/cpuinfo"; then variant1=vuln [ -z "$variant2" ] && variant2=immune variant3=vuln fi if [ "$capabilities_rdcl_no" = 1 ]; then # capability bit for future Intel processor that will explicitly state # that they're not vulnerable to Meltdown # this var is set in check_cpu() variant3=immune _debug "is_cpu_vulnerable: RDCL_NO is set so not vuln to meltdown" fi elif [ "$cpu_vendor" = AuthenticAMD ]; then # AMD revised their statement about variant2 => vulnerable # https://www.amd.com/en/corporate/speculative-execution variant1=vuln variant2=vuln [ -z "$variant3" ] && variant3=immune elif [ "$cpu_vendor" = ARM ]; then # ARM # reference: https://developer.arm.com/support/security-update # some devices (phones or other) have several ARMs and as such different part numbers, # an example is "bigLITTLE". we shouldn't rely on the first CPU only, so we check the whole list i=0 for cpupart in $cpu_part_list do i=$(( i + 1 )) # do NOT quote $cpu_arch_list below # shellcheck disable=SC2086 cpuarch=$(echo $cpu_arch_list | awk '{ print $'$i' }') _debug "checking cpu$i: <$cpupart> <$cpuarch>" # some kernels report AArch64 instead of 8 [ "$cpuarch" = "AArch64" ] && cpuarch=8 if [ -n "$cpupart" ] && [ -n "$cpuarch" ]; then # Cortex-R7 and Cortex-R8 are real-time and only used in medical devices or such # I can't find their CPU part number, but it's probably not that useful anyway # model R7 R8 A9 A15 A17 A57 A72 A73 A75 # part ? ? 0xc09 0xc0f 0xc0e 0xd07 0xd08 0xd09 0xd0a # arch 7? 7? 7 7 7 8 8 8 8 # # variant 1 & variant 2 if [ "$cpuarch" = 7 ] && echo "$cpupart" | grep -Eq '^0x(c09|c0f|c0e)$'; then # armv7 vulnerable chips _debug "checking cpu$i: this armv7 vulnerable to spectre 1 & 2" variant1=vuln variant2=vuln elif [ "$cpuarch" = 8 ] && echo "$cpupart" | grep -Eq '^0x(d07|d08|d09|d0a)$'; then # armv8 vulnerable chips _debug "checking cpu$i: this armv8 vulnerable to spectre 1 & 2" variant1=vuln variant2=vuln else _debug "checking cpu$i: this arm non vulnerable to 1 & 2" # others are not vulnerable [ -z "$variant1" ] && variant1=immune [ -z "$variant2" ] && variant2=immune fi # for variant3, only A75 is vulnerable if [ "$cpuarch" = 8 ] && [ "$cpupart" = 0xd0a ]; then _debug "checking cpu$i: arm A75 vulnerable to meltdown" variant3=vuln else _debug "checking cpu$i: this arm non vulnerable to meltdown" [ -z "$variant3" ] && variant3=immune fi fi _debug "is_cpu_vulnerable: for cpu$i and so far, we have <$variant1> <$variant2> <$variant3>" done fi _debug "is_cpu_vulnerable: temp results are <$variant1> <$variant2> <$variant3>" # if at least one of the cpu is vulnerable, then the system is vulnerable [ "$variant1" = "immune" ] && variant1=1 || variant1=0 [ "$variant2" = "immune" ] && variant2=1 || variant2=0 [ "$variant3" = "immune" ] && variant3=1 || variant3=0 _debug "is_cpu_vulnerable: final results are <$variant1> <$variant2> <$variant3>" is_cpu_vulnerable_cached=1 _is_cpu_vulnerable_cached "$1" return $? } is_cpu_specex_free() { # return true (0) if the CPU doesn't do speculative execution, false (1) if it does. # if it's not in the list we know, return false (1). # source: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/x86/kernel/cpu/common.c#n882 # { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_CEDARVIEW, X86_FEATURE_ANY }, # { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_CLOVERVIEW, X86_FEATURE_ANY }, # { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_LINCROFT, X86_FEATURE_ANY }, # { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_PENWELL, X86_FEATURE_ANY }, # { X86_VENDOR_INTEL, 6, INTEL_FAM6_ATOM_PINEVIEW, X86_FEATURE_ANY }, # { X86_VENDOR_CENTAUR, 5 }, # { X86_VENDOR_INTEL, 5 }, # { X86_VENDOR_NSC, 5 }, # { X86_VENDOR_ANY, 4 }, parse_cpu_details if [ "$cpu_vendor" = GenuineIntel ]; then if [ "$cpu_family" = 6 ]; then if [ "$cpu_model" = "$INTEL_FAM6_ATOM_CEDARVIEW" ] || \ [ "$cpu_model" = "$INTEL_FAM6_ATOM_CLOVERVIEW" ] || \ [ "$cpu_model" = "$INTEL_FAM6_ATOM_LINCROFT" ] || \ [ "$cpu_model" = "$INTEL_FAM6_ATOM_PENWELL" ] || \ [ "$cpu_model" = "$INTEL_FAM6_ATOM_PINEVIEW" ]; then return 0 fi elif [ "$cpu_family" = 5 ]; then return 0 fi fi [ "$cpu_family" = 4 ] && return 0 return 1 } show_header() { _info "Spectre and Meltdown mitigation detection tool v$VERSION" _info } parse_opt_file() { # parse_opt_file option_name option_value option_name="$1" option_value="$2" if [ -z "$option_value" ]; then show_header show_usage echo "$0: error: --$option_name expects one parameter (a file)" >&2 exit 1 elif [ ! -e "$option_value" ]; then show_header echo "$0: error: couldn't find file $option_value" >&2 exit 1 elif [ ! -f "$option_value" ]; then show_header echo "$0: error: $option_value is not a file" >&2 exit 1 elif [ ! -r "$option_value" ]; then show_header echo "$0: error: couldn't read $option_value (are you root?)" >&2 exit 1 fi echo "$option_value" exit 0 } while [ -n "$1" ]; do if [ "$1" = "--kernel" ]; then opt_kernel=$(parse_opt_file kernel "$2"); ret=$? [ $ret -ne 0 ] && exit 255 shift 2 opt_live=0 elif [ "$1" = "--config" ]; then opt_config=$(parse_opt_file config "$2"); ret=$? [ $ret -ne 0 ] && exit 255 shift 2 opt_live=0 elif [ "$1" = "--map" ]; then opt_map=$(parse_opt_file map "$2"); ret=$? [ $ret -ne 0 ] && exit 255 shift 2 opt_live=0 elif [ "$1" = "--arch-prefix" ]; then opt_arch_prefix="$2" shift 2 elif [ "$1" = "--live" ]; then opt_live_explicit=1 shift elif [ "$1" = "--no-color" ]; then opt_no_color=1 shift elif [ "$1" = "--no-sysfs" ]; then opt_no_sysfs=1 shift elif [ "$1" = "--sysfs-only" ]; then opt_sysfs_only=1 shift elif [ "$1" = "--coreos" ]; then opt_coreos=1 shift elif [ "$1" = "--coreos-within-toolbox" ]; then # don't use directly: used internally by --coreos opt_coreos=0 shift elif [ "$1" = "--hw-only" ]; then opt_hw_only=1 shift elif [ "$1" = "--no-hw" ]; then opt_no_hw=1 shift elif [ "$1" = "--no-explain" ]; then opt_no_explain=1 shift elif [ "$1" = "--batch" ]; then opt_batch=1 opt_verbose=0 shift case "$1" in text|nrpe|json|prometheus) opt_batch_format="$1"; shift;; --*) ;; # allow subsequent flags '') ;; # allow nothing at all *) echo "$0: error: unknown batch format '$1'" >&2 echo "$0: error: --batch expects a format from: text, nrpe, json" >&2 exit 255 ;; esac elif [ "$1" = "-v" ] || [ "$1" = "--verbose" ]; then opt_verbose=$(( opt_verbose + 1 )) shift elif [ "$1" = "--variant" ]; then if [ -z "$2" ]; then echo "$0: error: option --variant expects a parameter (1, 2 or 3)" >&2 exit 255 fi case "$2" in 1) opt_variant1=1; opt_allvariants=0;; 2) opt_variant2=1; opt_allvariants=0;; 3) opt_variant3=1; opt_allvariants=0;; *) echo "$0: error: invalid parameter '$2' for --variant, expected either 1, 2 or 3" >&2; exit 255 ;; esac shift 2 elif [ "$1" = "-h" ] || [ "$1" = "--help" ]; then show_header show_usage exit 0 elif [ "$1" = "--version" ]; then opt_no_color=1 show_header exit 0 elif [ "$1" = "--disclaimer" ]; then show_header show_disclaimer exit 0 else show_header show_usage echo "$0: error: unknown option '$1'" exit 255 fi done show_header if [ "$opt_no_sysfs" = 1 ] && [ "$opt_sysfs_only" = 1 ]; then _warn "Incompatible options specified (--no-sysfs and --sysfs-only), aborting" exit 255 fi if [ "$opt_no_hw" = 1 ] && [ "$opt_hw_only" = 1 ]; then _warn "Incompatible options specified (--no-hw and --hw-only), aborting" exit 255 fi # print status function pstatus() { if [ "$opt_no_color" = 1 ]; then _info_nol "$2" else case "$1" in red) col="\033[41m\033[30m";; green) col="\033[42m\033[30m";; yellow) col="\033[43m\033[30m";; blue) col="\033[44m\033[30m";; *) col="";; esac _info_nol "$col $2 \033[0m" fi [ -n "$3" ] && _info_nol " ($3)" _info unset col } # Print the final status of a vulnerability (incl. batch mode) # Arguments are: CVE UNK/OK/VULN description pvulnstatus() { if [ "$opt_batch" = 1 ]; then case "$1" in CVE-2017-5753) aka="SPECTRE VARIANT 1";; CVE-2017-5715) aka="SPECTRE VARIANT 2";; CVE-2017-5754) aka="MELTDOWN";; esac case "$opt_batch_format" in text) _echo 0 "$1: $2 ($3)";; json) case "$2" in UNK) is_vuln="null";; VULN) is_vuln="true";; OK) is_vuln="false";; esac json_output="${json_output:-[}{\"NAME\":\"$aka\",\"CVE\":\"$1\",\"VULNERABLE\":$is_vuln,\"INFOS\":\"$3\"}," ;; nrpe) [ "$2" = VULN ] && nrpe_vuln="$nrpe_vuln $1";; prometheus) prometheus_output="${prometheus_output:+$prometheus_output\n}specex_vuln_status{name=\"$aka\",cve=\"$1\",status=\"$2\",info=\"$3\"} 1" ;; esac fi # always fill global_* vars because we use that do decide the program exit code case "$2" in UNK) global_unknown="1";; VULN) global_critical="1";; esac # display info if we're not in quiet/batch mode vulnstatus="$2" shift 2 _info_nol "> \033[46m\033[30mSTATUS:\033[0m " case "$vulnstatus" in UNK) pstatus yellow 'UNKNOWN' "$@";; VULN) pstatus red 'VULNERABLE' "$@";; OK) pstatus green 'NOT VULNERABLE' "$@";; esac } # The 3 below functions are taken from the extract-linux script, available here: # https://github.com/torvalds/linux/blob/master/scripts/extract-vmlinux # The functions have been modified for better integration to this script # The original header of the file has been retained below # ---------------------------------------------------------------------- # extract-vmlinux - Extract uncompressed vmlinux from a kernel image # # Inspired from extract-ikconfig # (c) 2009,2010 Dick Streefland # # (c) 2011 Corentin Chary # # Licensed under the GNU General Public License, version 2 (GPLv2). # ---------------------------------------------------------------------- kernel='' kernel_err='' check_kernel() { _file="$1" _desperate_mode="$2" # checking the return code of readelf -h is not enough, we could get # a damaged ELF file and validate it, check for stderr warnings too _readelf_warnings=$("${opt_arch_prefix}readelf" -S "$_file" 2>&1 >/dev/null | tr "\n" "/"); ret=$? _readelf_sections=$("${opt_arch_prefix}readelf" -S "$_file" 2>/dev/null | grep -c -e data -e text -e init) _kernel_size=$(stat -c %s "$_file" 2>/dev/null || stat -f %z "$_file" 2>/dev/null || echo 10000) _debug "check_kernel: ret=$? size=$_kernel_size sections=$_readelf_sections warnings=$_readelf_warnings" if [ -n "$_desperate_mode" ]; then if "${opt_arch_prefix}strings" "$_file" | grep -Eq '^Linux version '; then _debug "check_kernel (desperate): ... matched!" return 0 else _debug "check_kernel (desperate): ... invalid" fi else if [ $ret -eq 0 ] && [ -z "$_readelf_warnings" ] && [ "$_readelf_sections" -gt 0 ]; then if [ "$_kernel_size" -ge 100000 ]; then _debug "check_kernel: ... file is valid" return 0 else _debug "check_kernel: ... file seems valid but is too small, ignoring" fi else _debug "check_kernel: ... file is invalid" fi fi return 1 } try_decompress() { # The obscure use of the "tr" filter is to work around older versions of # "grep" that report the byte offset of the line instead of the pattern. # Try to find the header ($1) and decompress from here _debug "try_decompress: looking for $3 magic in $6" for pos in $(tr "$1\n$2" "\n$2=" < "$6" | grep -abo "^$2") do _debug "try_decompress: magic for $3 found at offset $pos" if ! which "$3" >/dev/null 2>&1; then kernel_err="missing '$3' tool, please install it, usually it's in the '$5' package" return 0 fi pos=${pos%%:*} # shellcheck disable=SC2086 tail -c+$pos "$6" 2>/dev/null | $3 $4 > "$kerneltmp" 2>/dev/null; ret=$? if [ ! -s "$kerneltmp" ]; then # don't rely on $ret, sometimes it's != 0 but worked # (e.g. gunzip ret=2 just means there was trailing garbage) _debug "try_decompress: decompression with $3 failed (err=$ret)" elif check_kernel "$kerneltmp" "$7"; then kernel="$kerneltmp" _debug "try_decompress: decompressed with $3 successfully!" return 0 elif [ "$3" != "cat" ]; then _debug "try_decompress: decompression with $3 worked but result is not a kernel, trying with an offset" [ -z "$kerneltmp2" ] && kerneltmp2=$(mktemp /tmp/kernel-XXXXXX) cat "$kerneltmp" > "$kerneltmp2" try_decompress '\177ELF' xxy 'cat' '' cat "$kerneltmp2" && return 0 else _debug "try_decompress: decompression with $3 worked but result is not a kernel" fi done return 1 } extract_kernel() { [ -n "$1" ] || return 1 # Prepare temp files: kerneltmp="$(mktemp /tmp/kernel-XXXXXX)" # Initial attempt for uncompressed images or objects: if check_kernel "$1"; then cat "$1" > "$kerneltmp" kernel=$kerneltmp return 0 fi # That didn't work, so retry after decompression. for mode in '' 'desperate'; do try_decompress '\037\213\010' xy gunzip '' gunzip "$1" "$mode" && return 0 try_decompress '\3757zXZ\000' abcde unxz '' xz-utils "$1" "$mode" && return 0 try_decompress 'BZh' xy bunzip2 '' bzip2 "$1" "$mode" && return 0 try_decompress '\135\0\0\0' xxx unlzma '' xz-utils "$1" "$mode" && return 0 try_decompress '\211\114\132' xy 'lzop' '-d' lzop "$1" "$mode" && return 0 try_decompress '\002\041\114\030' xyy 'lz4' '-d -l' liblz4-tool "$1" "$mode" && return 0 try_decompress '\177ELF' xxy 'cat' '' cat "$1" "$mode" && return 0 done _verbose "Couldn't extract the kernel image, accuracy might be reduced" return 1 } # end of extract-vmlinux functions mount_debugfs() { if [ ! -e /sys/kernel/debug/sched_features ]; then # try to mount the debugfs hierarchy ourselves and remember it to umount afterwards mount -t debugfs debugfs /sys/kernel/debug 2>/dev/null && mounted_debugfs=1 fi } load_msr() { if [ "$os" = Linux ]; then modprobe msr 2>/dev/null && insmod_msr=1 _debug "attempted to load module msr, insmod_msr=$insmod_msr" else if ! kldstat -q -m cpuctl; then kldload cpuctl 2>/dev/null && kldload_cpuctl=1 _debug "attempted to load module cpuctl, kldload_cpuctl=$kldload_cpuctl" else _debug "cpuctl module already loaded" fi fi } load_cpuid() { if [ "$os" = Linux ]; then modprobe cpuid 2>/dev/null && insmod_cpuid=1 _debug "attempted to load module cpuid, insmod_cpuid=$insmod_cpuid" else if ! kldstat -q -m cpuctl; then kldload cpuctl 2>/dev/null && kldload_cpuctl=1 _debug "attempted to load module cpuctl, kldload_cpuctl=$kldload_cpuctl" else _debug "cpuctl module already loaded" fi fi } read_cpuid() { # leaf is the value of the eax register when calling the cpuid instruction: _leaf="$1" # eax=1 ebx=2 ecx=3 edx=4: _register="$2" # number of bits to shift the register right to: _shift="$3" # mask to apply as an AND operand to the shifted register value _mask="$4" # wanted value (optional), if present we return 0(true) if the obtained value is equal, 1 otherwise: _wanted="$5" # in any case, the read value is globally available in $read_cpuid_value read_cpuid_value='' if [ ! -e /dev/cpu/0/cpuid ] && [ ! -e /dev/cpuctl0 ]; then # try to load the module ourselves (and remember it so we can rmmod it afterwards) load_cpuid fi if [ -e /dev/cpu/0/cpuid ]; then # Linux # we need _leaf to be converted to decimal for dd _leaf=$(( _leaf )) _cpuid=$(dd if=/dev/cpu/0/cpuid bs=16 skip="$_leaf" iflag=skip_bytes count=1 2>/dev/null | od -A n -t u4) elif [ -e /dev/cpuctl0 ]; then # BSD _cpuid=$(cpucontrol -i "$_leaf" /dev/cpuctl0 2>/dev/null | awk '{print $4,$5,$6,$7}') # cpuid level 0x1: 0x000306d4 0x00100800 0x4dfaebbf 0xbfebfbff else return 2 fi _debug "cpuid: leaf$_leaf on cpu0, eax-ebx-ecx-edx: $_cpuid" [ -z "$_cpuid" ] && return 2 # get the value of the register we want _reg=$(echo "$_cpuid" | awk '{print $'"$_register"'}') # Linux returns it as decimal, BSD as hex, normalize to decimal _reg=$(( _reg )) # shellcheck disable=SC2046 _debug "cpuid: wanted register ($_register) has value $_reg aka "$(printf "%08x" "$_reg") _reg_shifted=$(( _reg >> _shift )) # shellcheck disable=SC2046 _debug "cpuid: shifted value by $_shift is $_reg_shifted aka "$(printf "%x" "$_reg_shifted") read_cpuid_value=$(( _reg_shifted & _mask )) # shellcheck disable=SC2046 _debug "cpuid: after AND $_mask, final value is $read_cpuid_value aka "$(printf "%x" "$read_cpuid_value") if [ -n "$_wanted" ]; then _debug "cpuid: wanted $_wanted and got $read_cpuid_value" if [ "$read_cpuid_value" = "$_wanted" ]; then return 0 else return 1 fi fi return 0 } dmesg_grep() { # grep for something in dmesg, ensuring that the dmesg buffer # has not been truncated dmesg_grepped='' if ! dmesg | grep -qE -e '(^|\] )Linux version [0-9]' -e '^FreeBSD is a registered' ; then # dmesg truncated return 2 fi dmesg_grepped=$(dmesg | grep -E "$1" | head -1) # not found: [ -z "$dmesg_grepped" ] && return 1 # found, output is in $dmesg_grepped return 0 } is_coreos() { which coreos-install >/dev/null 2>&1 && which toolbox >/dev/null 2>&1 && return 0 return 1 } parse_cpu_details() { [ "$parse_cpu_details_done" = 1 ] && return 0 if [ -e "$procfs/cpuinfo" ]; then cpu_vendor=$( grep '^vendor_id' "$procfs/cpuinfo" | awk '{print $3}' | head -1) cpu_friendly_name=$(grep '^model name' "$procfs/cpuinfo" | cut -d: -f2- | head -1 | sed -e 's/^ *//') # special case for ARM follows if grep -qi 'CPU implementer[[:space:]]*:[[:space:]]*0x41' "$procfs/cpuinfo"; then cpu_vendor='ARM' # some devices (phones or other) have several ARMs and as such different part numbers, # an example is "bigLITTLE", so we need to store the whole list, this is needed for is_cpu_vulnerable cpu_part_list=$(awk '/CPU part/ {print $4}' "$procfs/cpuinfo") cpu_arch_list=$(awk '/CPU architecture/ {print $3}' "$procfs/cpuinfo") # take the first one to fill the friendly name, do NOT quote the vars below # shellcheck disable=SC2086 cpu_arch=$(echo $cpu_arch_list | awk '{ print $1 }') # shellcheck disable=SC2086 cpu_part=$(echo $cpu_part_list | awk '{ print $1 }') [ "$cpu_arch" = "AArch64" ] && cpu_arch=8 cpu_friendly_name="ARM" [ -n "$cpu_arch" ] && cpu_friendly_name="$cpu_friendly_name v$cpu_arch" [ -n "$cpu_part" ] && cpu_friendly_name="$cpu_friendly_name model $cpu_part" fi cpu_family=$( grep '^cpu family' "$procfs/cpuinfo" | awk '{print $4}' | grep -E '^[0-9]+$' | head -1) cpu_model=$( grep '^model' "$procfs/cpuinfo" | awk '{print $3}' | grep -E '^[0-9]+$' | head -1) cpu_stepping=$(grep '^stepping' "$procfs/cpuinfo" | awk '{print $3}' | grep -E '^[0-9]+$' | head -1) cpu_ucode=$( grep '^microcode' "$procfs/cpuinfo" | awk '{print $3}' | head -1) else cpu_friendly_name=$(sysctl -n hw.model) fi # get raw cpuid, it's always useful (referenced in the Intel doc for firmware updates for example) if read_cpuid 0x1 1 0 0xFFFFFFFF; then cpuid="$read_cpuid_value" fi # under BSD, linprocfs often doesn't export ucode information, so fetch it ourselves the good old way if [ -z "$cpu_ucode" ] && [ "$os" != Linux ]; then load_cpuid if [ -e /dev/cpuctl0 ]; then # init MSR with NULLs cpucontrol -m 0x8b=0 /dev/cpuctl0 # call CPUID cpucontrol -i 1 /dev/cpuctl0 >/dev/null # read MSR cpu_ucode=$(cpucontrol -m 0x8b /dev/cpuctl0 | awk '{print $3}') # convert to decimal cpu_ucode=$(( cpu_ucode )) # convert back to hex cpu_ucode=$(printf "0x%x" "$cpu_ucode") fi fi echo "$cpu_ucode" | grep -q ^0x && cpu_ucode_decimal=$(( cpu_ucode )) ucode_found="model $cpu_model stepping $cpu_stepping ucode $cpu_ucode cpuid "$(printf "0x%x" "$cpuid") # also define those that we will need in other funcs # taken from ttps://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/x86/include/asm/intel-family.h # shellcheck disable=SC2034 { INTEL_FAM6_CORE_YONAH=$(( 0x0E )) INTEL_FAM6_CORE2_MEROM=$(( 0x0F )) INTEL_FAM6_CORE2_MEROM_L=$(( 0x16 )) INTEL_FAM6_CORE2_PENRYN=$(( 0x17 )) INTEL_FAM6_CORE2_DUNNINGTON=$(( 0x1D )) INTEL_FAM6_NEHALEM=$(( 0x1E )) INTEL_FAM6_NEHALEM_G=$(( 0x1F )) INTEL_FAM6_NEHALEM_EP=$(( 0x1A )) INTEL_FAM6_NEHALEM_EX=$(( 0x2E )) INTEL_FAM6_WESTMERE=$(( 0x25 )) INTEL_FAM6_WESTMERE_EP=$(( 0x2C )) INTEL_FAM6_WESTMERE_EX=$(( 0x2F )) INTEL_FAM6_SANDYBRIDGE=$(( 0x2A )) INTEL_FAM6_SANDYBRIDGE_X=$(( 0x2D )) INTEL_FAM6_IVYBRIDGE=$(( 0x3A )) INTEL_FAM6_IVYBRIDGE_X=$(( 0x3E )) INTEL_FAM6_HASWELL_CORE=$(( 0x3C )) INTEL_FAM6_HASWELL_X=$(( 0x3F )) INTEL_FAM6_HASWELL_ULT=$(( 0x45 )) INTEL_FAM6_HASWELL_GT3E=$(( 0x46 )) INTEL_FAM6_BROADWELL_CORE=$(( 0x3D )) INTEL_FAM6_BROADWELL_GT3E=$(( 0x47 )) INTEL_FAM6_BROADWELL_X=$(( 0x4F )) INTEL_FAM6_BROADWELL_XEON_D=$(( 0x56 )) INTEL_FAM6_SKYLAKE_MOBILE=$(( 0x4E )) INTEL_FAM6_SKYLAKE_DESKTOP=$(( 0x5E )) INTEL_FAM6_SKYLAKE_X=$(( 0x55 )) INTEL_FAM6_KABYLAKE_MOBILE=$(( 0x8E )) INTEL_FAM6_KABYLAKE_DESKTOP=$(( 0x9E )) # /* "Small Core" Processors (Atom) */ INTEL_FAM6_ATOM_PINEVIEW=$(( 0x1C )) INTEL_FAM6_ATOM_LINCROFT=$(( 0x26 )) INTEL_FAM6_ATOM_PENWELL=$(( 0x27 )) INTEL_FAM6_ATOM_CLOVERVIEW=$(( 0x35 )) INTEL_FAM6_ATOM_CEDARVIEW=$(( 0x36 )) INTEL_FAM6_ATOM_SILVERMONT1=$(( 0x37 )) INTEL_FAM6_ATOM_SILVERMONT2=$(( 0x4D )) INTEL_FAM6_ATOM_AIRMONT=$(( 0x4C )) INTEL_FAM6_ATOM_MERRIFIELD=$(( 0x4A )) INTEL_FAM6_ATOM_MOOREFIELD=$(( 0x5A )) INTEL_FAM6_ATOM_GOLDMONT=$(( 0x5C )) INTEL_FAM6_ATOM_DENVERTON=$(( 0x5F )) INTEL_FAM6_ATOM_GEMINI_LAKE=$(( 0x7A )) # /* Xeon Phi */ INTEL_FAM6_XEON_PHI_KNL=$(( 0x57 )) INTEL_FAM6_XEON_PHI_KNM=$(( 0x85 )) } parse_cpu_details_done=1 } is_ucode_blacklisted() { parse_cpu_details # if it's not an Intel, don't bother: it's not blacklisted [ "$cpu_vendor" = GenuineIntel ] || return 1 # it also needs to be family=6 [ "$cpu_family" = 6 ] || return 1 # now, check each known bad microcode # source: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/x86/kernel/cpu/intel.c#n105 # 2018-02-08 update: https://newsroom.intel.com/wp-content/uploads/sites/11/2018/02/microcode-update-guidance.pdf # model,stepping,microcode for tuple in \ $INTEL_FAM6_KABYLAKE_DESKTOP,0x0B,0x80 \ $INTEL_FAM6_KABYLAKE_DESKTOP,0x0A,0x80 \ $INTEL_FAM6_KABYLAKE_DESKTOP,0x09,0x80 \ $INTEL_FAM6_KABYLAKE_MOBILE,0x0A,0x80 \ $INTEL_FAM6_KABYLAKE_MOBILE,0x09,0x80 \ $INTEL_FAM6_SKYLAKE_X,0x03,0x0100013e \ $INTEL_FAM6_SKYLAKE_X,0x04,0x02000036 \ $INTEL_FAM6_SKYLAKE_X,0x04,0x0200003a \ $INTEL_FAM6_SKYLAKE_X,0x04,0x0200003c \ $INTEL_FAM6_BROADWELL_CORE,0x04,0x28 \ $INTEL_FAM6_BROADWELL_GT3E,0x01,0x1b \ $INTEL_FAM6_BROADWELL_XEON_D,0x02,0x14 \ $INTEL_FAM6_BROADWELL_XEON_D,0x03,0x07000011 \ $INTEL_FAM6_BROADWELL_X,0x01,0x0b000023 \ $INTEL_FAM6_BROADWELL_X,0x01,0x0b000025 \ $INTEL_FAM6_HASWELL_ULT,0x01,0x21 \ $INTEL_FAM6_HASWELL_GT3E,0x01,0x18 \ $INTEL_FAM6_HASWELL_CORE,0x03,0x23 \ $INTEL_FAM6_HASWELL_X,0x02,0x3b \ $INTEL_FAM6_HASWELL_X,0x04,0x10 \ $INTEL_FAM6_IVYBRIDGE_X,0x04,0x42a \ $INTEL_FAM6_SANDYBRIDGE_X,0x06,0x61b \ $INTEL_FAM6_SANDYBRIDGE_X,0x07,0x712 do model=$(echo $tuple | cut -d, -f1) stepping=$(( $(echo $tuple | cut -d, -f2) )) ucode=$(echo $tuple | cut -d, -f3) echo "$ucode" | grep -q ^0x && ucode_decimal=$(( ucode )) if [ "$cpu_model" = "$model" ] && [ "$cpu_stepping" = "$stepping" ]; then if [ "$cpu_ucode_decimal" = "$ucode_decimal" ] || [ "$cpu_ucode" = "$ucode" ]; then _debug "is_ucode_blacklisted: we have a match! ($cpu_model/$cpu_stepping/$cpu_ucode)" return 0 fi fi done _debug "is_ucode_blacklisted: no ($cpu_model/$cpu_stepping/$cpu_ucode)" return 1 } is_skylake_cpu() { # is this a skylake cpu? # return 0 if yes, 1 otherwise #if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL && # boot_cpu_data.x86 == 6) { # switch (boot_cpu_data.x86_model) { # case INTEL_FAM6_SKYLAKE_MOBILE: # case INTEL_FAM6_SKYLAKE_DESKTOP: # case INTEL_FAM6_SKYLAKE_X: # case INTEL_FAM6_KABYLAKE_MOBILE: # case INTEL_FAM6_KABYLAKE_DESKTOP: # return true; parse_cpu_details [ "$cpu_vendor" = GenuineIntel ] || return 1 [ "$cpu_family" = 6 ] || return 1 if [ "$cpu_model" = $INTEL_FAM6_SKYLAKE_MOBILE ] || \ [ "$cpu_model" = $INTEL_FAM6_SKYLAKE_DESKTOP ] || \ [ "$cpu_model" = $INTEL_FAM6_SKYLAKE_X ] || \ [ "$cpu_model" = $INTEL_FAM6_KABYLAKE_MOBILE ] || \ [ "$cpu_model" = $INTEL_FAM6_KABYLAKE_DESKTOP ]; then return 0 fi return 1 } # ENTRYPOINT # we can't do anything useful under WSL if uname -a | grep -qE -- '-Microsoft #[0-9]+-Microsoft '; then _warn "This script doesn't work under Windows Subsystem for Linux" _warn "You should use the official Microsoft tool instead." _warn "It can be found under https://aka.ms/SpeculationControlPS" exit 1 fi # check for mode selection inconsistency if [ "$opt_live_explicit" = 1 ]; then if [ -n "$opt_kernel" ] || [ -n "$opt_config" ] || [ -n "$opt_map" ]; then show_usage echo "$0: error: incompatible modes specified, use either --live or --kernel/--config/--map" >&2 exit 255 fi fi if [ "$opt_hw_only" = 1 ]; then if [ "$opt_allvariants" = 0 ]; then show_usage echo "$0: error: incompatible modes specified, --hw-only vs --variant" >&2 exit 255 else opt_allvariants=0 opt_variant1=0 opt_variant2=0 opt_variant3=0 fi fi # coreos mode if [ "$opt_coreos" = 1 ]; then if ! is_coreos; then _warn "CoreOS mode asked, but we're not under CoreOS!" exit 255 fi _warn "CoreOS mode, starting an ephemeral toolbox to launch the script" load_msr load_cpuid mount_debugfs toolbox --ephemeral --bind-ro /dev/cpu:/dev/cpu -- sh -c "dnf install -y binutils which && /media/root$PWD/$0 $* --coreos-within-toolbox" exitcode=$? exit $exitcode else if is_coreos; then _warn "You seem to be running CoreOS, you might want to use the --coreos option for better results" _warn fi fi # if we're under a BSD, try to mount linprocfs for "$procfs/cpuinfo" procfs=/proc if echo "$os" | grep -q BSD; then _debug "We're under BSD, check if we have procfs" procfs=$(mount | awk '/^linprocfs/ { print $3; exit; }') if [ -z "$procfs" ]; then _debug "we don't, try to mount it" procfs=/proc [ -d /compat/linux/proc ] && procfs=/compat/linux/proc test -d $procfs || mkdir $procfs if mount -t linprocfs linprocfs $procfs 2>/dev/null; then mounted_procfs=1 _debug "procfs just mounted at $procfs" else procfs='' fi else _debug "We do: $procfs" fi fi parse_cpu_details if [ "$opt_live" = 1 ]; then # root check (only for live mode, for offline mode, we already checked if we could read the files) if [ "$(id -u)" -ne 0 ]; then _warn "Note that you should launch this script with root privileges to get accurate information." _warn "We'll proceed but you might see permission denied errors." _warn "To run it as root, you can try the following command: sudo $0" _warn fi _info "Checking for vulnerabilities on current system" _info "Kernel is \033[35m$(uname -s) $(uname -r) $(uname -v) $(uname -m)\033[0m" _info "CPU is \033[35m$cpu_friendly_name\033[0m" # try to find the image of the current running kernel # first, look for the BOOT_IMAGE hint in the kernel cmdline if [ -r /proc/cmdline ] && grep -q 'BOOT_IMAGE=' /proc/cmdline; then opt_kernel=$(grep -Eo 'BOOT_IMAGE=[^ ]+' /proc/cmdline | cut -d= -f2) _debug "found opt_kernel=$opt_kernel in /proc/cmdline" # if we have a dedicated /boot partition, our bootloader might have just called it / # so try to prepend /boot and see if we find anything [ -e "/boot/$opt_kernel" ] && opt_kernel="/boot/$opt_kernel" # special case for CoreOS if we're inside the toolbox [ -e "/media/root/boot/$opt_kernel" ] && opt_kernel="/media/root/boot/$opt_kernel" _debug "opt_kernel is now $opt_kernel" # else, the full path is already there (most probably /boot/something) fi # if we didn't find a kernel, default to guessing if [ ! -e "$opt_kernel" ]; then # Fedora: [ -e "/lib/modules/$(uname -r)/vmlinuz" ] && opt_kernel="/lib/modules/$(uname -r)/vmlinuz" # Slackare: [ -e "/boot/vmlinuz" ] && opt_kernel="/boot/vmlinuz" # Arch: [ -e "/boot/vmlinuz-linux" ] && opt_kernel="/boot/vmlinuz-linux" # Linux-Libre: [ -e "/boot/vmlinuz-linux-libre" ] && opt_kernel="/boot/vmlinuz-linux-libre" # pine64 [ -e "/boot/pine64/Image" ] && opt_kernel="/boot/pine64/Image" # generic: [ -e "/boot/vmlinuz-$(uname -r)" ] && opt_kernel="/boot/vmlinuz-$(uname -r)" [ -e "/boot/kernel-$( uname -r)" ] && opt_kernel="/boot/kernel-$( uname -r)" [ -e "/boot/bzImage-$(uname -r)" ] && opt_kernel="/boot/bzImage-$(uname -r)" # Gentoo: [ -e "/boot/kernel-genkernel-$(uname -m)-$(uname -r)" ] && opt_kernel="/boot/kernel-genkernel-$(uname -m)-$(uname -r)" # NixOS: [ -e "/run/booted-system/kernel" ] && opt_kernel="/run/booted-system/kernel" # systemd kernel-install: [ -e "/etc/machine-id" ] && [ -e "/boot/$(cat /etc/machine-id)/$(uname -r)/linux" ] && opt_kernel="/boot/$(cat /etc/machine-id)/$(uname -r)/linux" fi # system.map if [ -e /proc/kallsyms ] ; then opt_map=/proc/kallsyms elif [ -e "/lib/modules/$(uname -r)/System.map" ] ; then opt_map="/lib/modules/$(uname -r)/System.map" elif [ -e "/boot/System.map-$(uname -r)" ] ; then opt_map="/boot/System.map-$(uname -r)" fi # config if [ -e /proc/config.gz ] ; then dumped_config="$(mktemp /tmp/config-XXXXXX)" gunzip -c /proc/config.gz > "$dumped_config" # dumped_config will be deleted at the end of the script opt_config="$dumped_config" elif [ -e "/lib/modules/$(uname -r)/config" ]; then opt_config="/lib/modules/$(uname -r)/config" elif [ -e "/boot/config-$(uname -r)" ]; then opt_config="/boot/config-$(uname -r)" fi else _info "Checking for vulnerabilities against specified kernel" _info "CPU is \033[35m$cpu_friendly_name\033[0m" fi if [ -n "$opt_kernel" ]; then _verbose "Will use kernel image \033[35m$opt_kernel\033[0m" else _verbose "Will use no kernel image (accuracy might be reduced)" bad_accuracy=1 fi if [ "$os" = Linux ]; then if [ -n "$opt_config" ] && ! grep -q '^CONFIG_' "$opt_config"; then # given file is invalid! _warn "The kernel config file seems invalid, was expecting a plain-text file, ignoring it!" opt_config='' fi if [ -n "$dumped_config" ] && [ -n "$opt_config" ]; then _verbose "Will use kconfig \033[35m/proc/config.gz (decompressed)\033[0m" elif [ -n "$opt_config" ]; then _verbose "Will use kconfig \033[35m$opt_config\033[0m" else _verbose "Will use no kconfig (accuracy might be reduced)" bad_accuracy=1 fi if [ -n "$opt_map" ]; then _verbose "Will use System.map file \033[35m$opt_map\033[0m" else _verbose "Will use no System.map file (accuracy might be reduced)" bad_accuracy=1 fi if [ "$bad_accuracy" = 1 ]; then _info "We're missing some kernel info (see -v), accuracy might be reduced" fi fi if [ -e "$opt_kernel" ]; then if ! which "${opt_arch_prefix}readelf" >/dev/null 2>&1; then _debug "readelf not found" kernel_err="missing '${opt_arch_prefix}readelf' tool, please install it, usually it's in the 'binutils' package" elif [ "$opt_sysfs_only" = 1 ]; then kernel_err='kernel image decompression skipped' else extract_kernel "$opt_kernel" fi else _debug "no opt_kernel defined" kernel_err="couldn't find your kernel image in /boot, if you used netboot, this is normal" fi if [ -z "$kernel" ] || [ ! -r "$kernel" ]; then [ -z "$kernel_err" ] && kernel_err="couldn't extract your kernel from $opt_kernel" else # vanilla kernels have with ^Linux version # also try harder with some kernels (such as Red Hat) that don't have ^Linux version before their version string # and check for FreeBSD kernel_version=$("${opt_arch_prefix}strings" "$kernel" 2>/dev/null | grep -E \ -e '^Linux version ' \ -e '^[[:alnum:]][^[:space:]]+ \([^[:space:]]+\) #[0-9]+ .+ (19|20)[0-9][0-9]$' \ -e '^FreeBSD [0-9]' | head -1) if [ -z "$kernel_version" ]; then # try even harder with some kernels (such as ARM) that split the release (uname -r) and version (uname -v) in 2 adjacent strings kernel_version=$("${opt_arch_prefix}strings" "$kernel" 2>/dev/null | grep -E -B1 '^#[0-9]+ .+ (19|20)[0-9][0-9]$' | tr "\n" " ") fi if [ -n "$kernel_version" ]; then # in live mode, check if the img we found is the correct one if [ "$opt_live" = 1 ]; then _verbose "Kernel image is \033[35m$kernel_version" if ! echo "$kernel_version" | grep -qF "$(uname -r)"; then _warn "Possible disrepancy between your running kernel '$(uname -r)' and the image '$kernel_version' we found ($opt_kernel), results might be incorrect" fi else _info "Kernel image is \033[35m$kernel_version" fi else _verbose "Kernel image version is unknown" fi fi _info # end of header stuff # now we define some util functions and the check_*() funcs, as # the user can choose to execute only some of those sys_interface_check() { [ "$opt_live" = 1 ] && [ "$opt_no_sysfs" = 0 ] && [ -r "$1" ] || return 1 _info_nol "* Mitigated according to the /sys interface: " if grep -qi '^not affected' "$1"; then # Not affected status=OK pstatus green YES "kernel confirms that your CPU is unaffected" elif grep -qi '^mitigation' "$1"; then # Mitigation: PTI status=OK pstatus green YES "kernel confirms that the mitigation is active" elif grep -qi '^vulnerable' "$1"; then # Vulnerable status=VULN pstatus yellow NO "kernel confirms your system is vulnerable" else status=UNK pstatus yellow UNKNOWN "unknown value reported by kernel" fi msg=$(cat "$1") _debug "sys_interface_check: $1=$msg" return 0 } number_of_cpus() { if echo "$os" | grep -q BSD; then n=$(sysctl -n hw.ncpu 2>/dev/null || echo 1) elif [ -e "$procfs/cpuinfo" ]; then n=$(grep -c ^processor "$procfs/cpuinfo" 2>/dev/null || echo 1) else # if we don't know, default to 1 CPU n=1 fi return "$n" } # $1 - msr number # $2 - cpu index write_msr() { if [ "$os" != Linux ]; then cpucontrol -m "$1=0" "/dev/cpuctl$2" >/dev/null 2>&1; ret=$? else # convert to decimal _msrindex=$(( $1 )) if [ ! -w /dev/cpu/"$2"/msr ]; then ret=200 # permission error else dd if=/dev/zero of=/dev/cpu/"$2"/msr bs=8 count=1 seek="$_msrindex" oflag=seek_bytes 2>/dev/null; ret=$? fi fi _debug "write_msr: for cpu $2 on msr $1 ($_msrindex), ret=$ret" return $ret } read_msr() { # _msr must be in hex, in the form 0x1234: _msr="$1" # cpu index, starting from 0: _cpu="$2" read_msr_value='' if [ "$os" != Linux ]; then _msr=$(cpucontrol -m "$_msr" "/dev/cpuctl$_cpu" 2>/dev/null); ret=$? [ $ret -ne 0 ] && return 1 # MSR 0x10: 0x000003e1 0xb106dded _msr_h=$(echo "$_msr" | awk '{print $3}'); _msr_h="$(( _msr_h >> 24 & 0xFF )) $(( _msr_h >> 16 & 0xFF )) $(( _msr_h >> 8 & 0xFF )) $(( _msr_h & 0xFF ))" _msr_l=$(echo "$_msr" | awk '{print $4}'); _msr_l="$(( _msr_l >> 24 & 0xFF )) $(( _msr_l >> 16 & 0xFF )) $(( _msr_l >> 8 & 0xFF )) $(( _msr_l & 0xFF ))" read_msr_value="$_msr_h $_msr_l" else # convert to decimal _msr=$(( _msr )) if [ ! -r /dev/cpu/"$_cpu"/msr ]; then return 200 # permission error fi read_msr_value=$(dd if=/dev/cpu/"$_cpu"/msr bs=8 count=1 skip="$_msr" iflag=skip_bytes 2>/dev/null | od -t u1 -A n) if [ -z "$read_msr_value" ]; then # MSR doesn't exist, don't check for $? because some versions of dd still return 0! return 1 fi fi _debug "read_msr: MSR=$1 value is $read_msr_value" return 0 } check_cpu() { _info "\033[1;34mHardware check\033[0m" if ! uname -m | grep -qwE 'x86_64|i[3-6]86|amd64'; then return fi _info "* Hardware support (CPU microcode) for mitigation techniques" _info " * Indirect Branch Restricted Speculation (IBRS)" _info_nol " * SPEC_CTRL MSR is available: " number_of_cpus ncpus=$? idx_max_cpu=$((ncpus-1)) if [ ! -e /dev/cpu/0/msr ] && [ ! -e /dev/cpuctl0 ]; then # try to load the module ourselves (and remember it so we can rmmod it afterwards) load_msr fi if [ ! -e /dev/cpu/0/msr ] && [ ! -e /dev/cpuctl0 ]; then spec_ctrl_msr=-1 pstatus yellow UNKNOWN "is msr kernel module available?" else # the new MSR 'SPEC_CTRL' is at offset 0x48 # here we use dd, it's the same as using 'rdmsr 0x48' but without needing the rdmsr tool # if we get a read error, the MSR is not there. bs has to be 8 for msr # skip=9 because 8*9=72=0x48 val=0 cpu_mismatch=0 for i in $(seq 0 "$idx_max_cpu") do read_msr 0x48 "$i"; ret=$? if [ "$i" -eq 0 ]; then val=$ret else if [ "$ret" -eq $val ]; then continue else cpu_mismatch=1 fi fi done if [ $val -eq 0 ]; then if [ $cpu_mismatch -eq 0 ]; then spec_ctrl_msr=1 pstatus green YES else spec_ctrl_msr=1 pstatus green YES "But not in all CPUs" fi elif [ $val -eq 200 ]; then pstatus yellow UNKNOWN "is msr kernel module available?" spec_ctrl_msr=-1 else spec_ctrl_msr=0 pstatus yellow NO fi fi _info_nol " * CPU indicates IBRS capability: " # from kernel src: { X86_FEATURE_SPEC_CTRL, CPUID_EDX,26, 0x00000007, 0 }, read_cpuid 0x7 4 26 1 1; ret=$? if [ $ret -eq 0 ]; then pstatus green YES "SPEC_CTRL feature bit" cpuid_spec_ctrl=1 elif [ $ret -eq 2 ]; then pstatus yellow UNKNOWN "is cpuid kernel module available?" else pstatus yellow NO fi # hardware support according to kernel if [ "$opt_verbose" -ge 2 ]; then # the spec_ctrl flag in cpuinfo is set if and only if the kernel sees # that the spec_ctrl cpuinfo bit set. we already check that ourselves above # but let's check it anyway (in verbose mode only) _verbose_nol " * Kernel has set the spec_ctrl flag in cpuinfo: " if [ "$opt_live" = 1 ]; then if grep ^flags "$procfs/cpuinfo" | grep -qw spec_ctrl; then pstatus blue YES else pstatus blue NO fi else pstatus blue N/A "not testable in offline mode" fi fi # IBPB _info " * Indirect Branch Prediction Barrier (IBPB)" _info_nol " * PRED_CMD MSR is available: " if [ ! -e /dev/cpu/0/msr ] && [ ! -e /dev/cpuctl0 ]; then pstatus yellow UNKNOWN "is msr kernel module available?" else # the new MSR 'PRED_CTRL' is at offset 0x49, write-only # here we use dd, it's the same as using 'wrmsr 0x49 0' but without needing the wrmsr tool # if we get a write error, the MSR is not there val=0 cpu_mismatch=0 for i in $(seq 0 "$idx_max_cpu") do write_msr 0x49 "$i"; ret=$? if [ "$i" -eq 0 ]; then val=$ret else if [ "$ret" -eq $val ]; then continue else cpu_mismatch=1 fi fi done if [ $val -eq 0 ]; then if [ $cpu_mismatch -eq 0 ]; then pstatus green YES else pstatus green YES "But not in all CPUs" fi elif [ $val -eq 200 ]; then pstatus yellow UNKNOWN "is msr kernel module available?" else pstatus yellow NO fi fi _info_nol " * CPU indicates IBPB capability: " # CPUID EAX=0x80000008, ECX=0x00 return EBX[12] indicates support for just IBPB. read_cpuid 0x80000008 2 12 1 1; ret=$? if [ $ret -eq 0 ]; then pstatus green YES "IBPB_SUPPORT feature bit" elif [ "$cpuid_spec_ctrl" = 1 ]; then pstatus green YES "SPEC_CTRL feature bit" elif [ $ret -eq 2 ]; then pstatus yellow UNKNOWN "is cpuid kernel module available?" else pstatus yellow NO fi # STIBP _info " * Single Thread Indirect Branch Predictors (STIBP)" _info_nol " * SPEC_CTRL MSR is available: " if [ "$spec_ctrl_msr" = 1 ]; then pstatus green YES elif [ "$spec_ctrl_msr" = 0 ]; then pstatus yellow NO else pstatus yellow UNKNOWN "is msr kernel module available?" fi _info_nol " * CPU indicates STIBP capability: " # A processor supports STIBP if it enumerates CPUID (EAX=7H,ECX=0):EDX[27] as 1 read_cpuid 0x7 4 27 1 1; ret=$? if [ $ret -eq 0 ]; then pstatus green YES elif [ $ret -eq 2 ]; then pstatus yellow UNKNOWN "is cpuid kernel module available?" else pstatus yellow NO fi _info " * Enhanced IBRS (IBRS_ALL)" _info_nol " * CPU indicates ARCH_CAPABILITIES MSR availability: " cpuid_arch_capabilities=-1 # A processor supports the ARCH_CAPABILITIES MSR if it enumerates CPUID (EAX=7H,ECX=0):EDX[29] as 1 read_cpuid 0x7 4 29 1 1; ret=$? if [ $ret -eq 0 ]; then pstatus green YES cpuid_arch_capabilities=1 elif [ $ret -eq 2 ]; then pstatus yellow UNKNOWN "is cpuid kernel module available?" else pstatus yellow NO cpuid_arch_capabilities=0 fi _info_nol " * ARCH_CAPABILITIES MSR advertises IBRS_ALL capability: " capabilities_rdcl_no=-1 capabilities_ibrs_all=-1 if [ "$cpuid_arch_capabilities" = -1 ]; then pstatus yellow UNKNOWN elif [ "$cpuid_arch_capabilities" != 1 ]; then capabilities_rdcl_no=0 capabilities_ibrs_all=0 pstatus yellow NO elif [ ! -e /dev/cpu/0/msr ] && [ ! -e /dev/cpuctl0 ]; then spec_ctrl_msr=-1 pstatus yellow UNKNOWN "is msr kernel module available?" else # the new MSR 'ARCH_CAPABILITIES' is at offset 0x10a # here we use dd, it's the same as using 'rdmsr 0x10a' but without needing the rdmsr tool # if we get a read error, the MSR is not there. bs has to be 8 for msr val=0 val_cap_msr=0 cpu_mismatch=0 for i in $(seq 0 "$idx_max_cpu") do read_msr 0x10a "$i"; ret=$? capabilities=$(echo "$read_msr_value" | awk '{print $8}') if [ "$i" -eq 0 ]; then val=$ret val_cap_msr=$capabilities else if [ "$ret" -eq "$val" ] && [ "$capabilities" -eq "$val_cap_msr" ]; then continue else cpu_mismatch=1 fi fi done capabilities=$val_cap_msr capabilities_rdcl_no=0 capabilities_ibrs_all=0 if [ $val -eq 0 ]; then _debug "capabilities MSR lower byte is $capabilities (decimal)" [ $(( capabilities & 1 )) -eq 1 ] && capabilities_rdcl_no=1 [ $(( capabilities & 2 )) -eq 2 ] && capabilities_ibrs_all=1 _debug "capabilities says rdcl_no=$capabilities_rdcl_no ibrs_all=$capabilities_ibrs_all" if [ "$capabilities_ibrs_all" = 1 ]; then if [ $cpu_mismatch -eq 0 ]; then pstatus green YES else: pstatus green YES "But not in all CPUs" fi else pstatus yellow NO fi elif [ $val -eq 200 ]; then pstatus yellow UNKNOWN "is msr kernel module available?" else pstatus yellow NO fi fi _info_nol " * CPU explicitly indicates not being vulnerable to Meltdown (RDCL_NO): " if [ "$capabilities_rdcl_no" = -1 ]; then pstatus yellow UNKNOWN elif [ "$capabilities_rdcl_no" = 1 ]; then pstatus green YES else pstatus yellow NO fi _info_nol " * CPU microcode is known to cause stability problems: " if is_ucode_blacklisted; then pstatus red YES "$ucode_found" _warn _warn "The microcode your CPU is running on is known to cause instability problems," _warn "such as intempestive reboots or random crashes." _warn "You are advised to either revert to a previous microcode version (that might not have" _warn "the mitigations for Spectre), or upgrade to a newer one if available." _warn else pstatus blue NO "$ucode_found" fi } check_cpu_vulnerabilities() { _info "* CPU vulnerability to the three speculative execution attack variants" for v in 1 2 3; do _info_nol " * Vulnerable to Variant $v: " if is_cpu_vulnerable $v; then pstatus yellow YES else pstatus green NO fi done } check_redhat_canonical_spectre() { # if we were already called, don't do it again [ -n "$redhat_canonical_spectre" ] && return if ! which "${opt_arch_prefix}strings" >/dev/null 2>&1; then redhat_canonical_spectre=-1 elif [ -n "$kernel_err" ]; then redhat_canonical_spectre=-2 else # Red Hat / Ubuntu specific variant1 patch is difficult to detect, # let's use the two same tricks than the official Red Hat detection script uses: if "${opt_arch_prefix}strings" "$kernel" | grep -qw noibrs && "${opt_arch_prefix}strings" "$kernel" | grep -qw noibpb; then # 1) detect their specific variant2 patch. If it's present, it means # that the variant1 patch is also present (both were merged at the same time) _debug "found redhat/canonical version of the variant2 patch (implies variant1)" redhat_canonical_spectre=1 elif "${opt_arch_prefix}strings" "$kernel" | grep -q 'x86/pti:'; then # 2) detect their specific variant3 patch. If it's present, but the variant2 # is not, it means that only variant1 is present in addition to variant3 _debug "found redhat/canonical version of the variant3 patch (implies variant1 but not variant2)" redhat_canonical_spectre=2 else redhat_canonical_spectre=0 fi fi } ################### # SPECTRE VARIANT 1 check_variant1() { _info "\033[1;34mCVE-2017-5753 [bounds check bypass] aka 'Spectre Variant 1'\033[0m" if [ "$os" = Linux ]; then check_variant1_linux elif echo "$os" | grep -q BSD; then check_variant1_bsd else _warn "Unsupported OS ($os)" fi } check_variant1_linux() { status=UNK sys_interface_available=0 msg='' if sys_interface_check "/sys/devices/system/cpu/vulnerabilities/spectre_v1"; then # this kernel has the /sys interface, trust it over everything # v0.33+: don't. some kernels have backported the array_index_mask_nospec() workaround without # modifying the vulnerabilities/spectre_v1 file. that's bad. we can't trust it when it says Vulnerable :( # see "silent backport" detection at the bottom of this func sys_interface_available=1 fi if [ "$opt_sysfs_only" != 1 ]; then # no /sys interface (or offline mode), fallback to our own ways _info_nol "* Kernel has array_index_mask_nospec (x86): " # vanilla: look for the Linus' mask aka array_index_mask_nospec() # that is inlined at least in raw_copy_from_user (__get_user_X symbols) #mov PER_CPU_VAR(current_task), %_ASM_DX #cmp TASK_addr_limit(%_ASM_DX),%_ASM_AX #jae bad_get_user # /* array_index_mask_nospec() are the 2 opcodes that follow */ #+sbb %_ASM_DX, %_ASM_DX #+and %_ASM_DX, %_ASM_AX #ASM_STAC # x86 64bits: jae(0x0f 0x83 0x?? 0x?? 0x?? 0x??) sbb(0x48 0x19 0xd2) and(0x48 0x21 0xd0) # x86 32bits: cmp(0x3b 0x82 0x?? 0x?? 0x00 0x00) jae(0x73 0x??) sbb(0x19 0xd2) and(0x21 0xd0) if [ -n "$kernel_err" ]; then pstatus yellow UNKNOWN "couldn't check ($kernel_err)" elif ! which perl >/dev/null 2>&1; then pstatus yellow UNKNOWN "missing 'perl' binary, please install it" else perl -ne '/\x0f\x83....\x48\x19\xd2\x48\x21\xd0/ and $found++; END { exit($found) }' "$kernel"; ret=$? if [ $ret -gt 0 ]; then pstatus green YES "$ret occurrence(s) found of 64 bits array_index_mask_nospec()" v1_mask_nospec="64 bits array_index_mask_nospec" else perl -ne '/\x3b\x82..\x00\x00\x73.\x19\xd2\x21\xd0/ and $found++; END { exit($found) }' "$kernel"; ret=$? if [ $ret -gt 0 ]; then pstatus green YES "$ret occurrence(s) found of 32 bits array_index_mask_nospec()" v1_mask_nospec="32 bits array_index_mask_nospec" else pstatus yellow NO fi fi fi _info_nol "* Kernel has the Red Hat/Ubuntu patch: " check_redhat_canonical_spectre if [ "$redhat_canonical_spectre" = -1 ]; then pstatus yellow UNKNOWN "missing '${opt_arch_prefix}strings' tool, please install it, usually it's in the binutils package" elif [ "$redhat_canonical_spectre" = -2 ]; then pstatus yellow UNKNOWN "couldn't check ($kernel_err)" elif [ "$redhat_canonical_spectre" = 1 ]; then pstatus green YES elif [ "$redhat_canonical_spectre" = 2 ]; then pstatus green YES "but without IBRS" else pstatus yellow NO fi _info_nol "* Kernel has mask_nospec64 (arm): " #.macro mask_nospec64, idx, limit, tmp #sub \tmp, \idx, \limit #bic \tmp, \tmp, \idx #and \idx, \idx, \tmp, asr #63 #csdb #.endm #$ aarch64-linux-gnu-objdump -d vmlinux | grep -w bic -A1 -B1 | grep -w sub -A2 | grep -w and -B2 #ffffff8008082e44: cb190353 sub x19, x26, x25 #ffffff8008082e48: 8a3a0273 bic x19, x19, x26 #ffffff8008082e4c: 8a93ff5a and x26, x26, x19, asr #63 #ffffff8008082e50: d503229f hint #0x14 # if we have v1_mask_nospec or redhat_canonical_spectre>0, don't bother disassembling the kernel, the answer is no. if [ -n "$v1_mask_nospec" ] || [ "$redhat_canonical_spectre" -gt 0 ]; then pstatus yellow NO elif [ -n "$kernel_err" ]; then pstatus yellow UNKNOWN "couldn't check ($kernel_err)" elif ! which perl >/dev/null 2>&1; then pstatus yellow UNKNOWN "missing 'perl' binary, please install it" elif ! which "${opt_arch_prefix}objdump" >/dev/null 2>&1; then pstatus yellow UNKNOWN "missing '${opt_arch_prefix}objdump' tool, please install it, usually it's in the binutils package" else "${opt_arch_prefix}objdump" -d "$kernel" | perl -ne 'push @r, $_; /\shint\s/ && $r[0]=~/\ssub\s+(x\d+)/ && $r[1]=~/\sbic\s+$1,\s+$1,/ && $r[2]=~/\sand\s/ && exit(9); shift @r if @r>3'; ret=$? if [ "$ret" -eq 9 ]; then pstatus green YES "mask_nospec64 macro is present and used" v1_mask_nospec="arm mask_nospec64" else pstatus yellow NO fi fi if [ "$opt_verbose" -ge 2 ] || ( [ -z "$v1_mask_nospec" ] && [ "$redhat_canonical_spectre" != 1 ] && [ "$redhat_canonical_spectre" != 2 ] ); then # this is a slow heuristic and we don't need it if we already know the kernel is patched # but still show it in verbose mode _info_nol "* Checking count of LFENCE instructions following a jump in kernel... " if [ -n "$kernel_err" ]; then pstatus yellow UNKNOWN "couldn't check ($kernel_err)" else if ! which "${opt_arch_prefix}objdump" >/dev/null 2>&1; then pstatus yellow UNKNOWN "missing '${opt_arch_prefix}objdump' tool, please install it, usually it's in the binutils package" else # here we disassemble the kernel and count the number of occurrences of the LFENCE opcode # in non-patched kernels, this has been empirically determined as being around 40-50 # in patched kernels, this is more around 70-80, sometimes way higher (100+) # v0.13: 68 found in a 3.10.23-xxxx-std-ipv6-64 (with lots of modules compiled-in directly), which doesn't have the LFENCE patches, # so let's push the threshold to 70. # v0.33+: now only count lfence opcodes after a jump, way less error-prone # non patched kernel have between 0 and 20 matches, patched ones have at least 40-45 nb_lfence=$("${opt_arch_prefix}objdump" -d "$kernel" 2>/dev/null | grep -w -B1 lfence | grep -Ewc 'jmp|jne|je') if [ "$nb_lfence" -lt 30 ]; then pstatus yellow NO "only $nb_lfence jump-then-lfence instructions found, should be >= 30 (heuristic)" else v1_lfence=1 pstatus green YES "$nb_lfence jump-then-lfence instructions found, which is >= 30 (heuristic)" fi fi fi fi else # we have no sysfs but were asked to use it only! msg="/sys vulnerability interface use forced, but it's not available!" status=UNK fi # report status cve='CVE-2017-5753' if ! is_cpu_vulnerable 1; then # override status & msg in case CPU is not vulnerable after all pvulnstatus $cve OK "your CPU vendor reported your CPU model as not vulnerable" elif [ -z "$msg" ]; then # if msg is empty, sysfs check didn't fill it, rely on our own test if [ -n "$v1_mask_nospec" ]; then pvulnstatus $cve OK "Kernel source has been patched to mitigate the vulnerability ($v1_mask_nospec)" elif [ "$redhat_canonical_spectre" = 1 ] || [ "$redhat_canonical_spectre" = 2 ]; then pvulnstatus $cve OK "Kernel source has been patched to mitigate the vulnerability (Red Hat/Ubuntu patch)" elif [ "$v1_lfence" = 1 ]; then pvulnstatus $cve OK "Kernel source has PROBABLY been patched to mitigate the vulnerability (jump-then-lfence instructions heuristic)" elif [ "$kernel_err" ]; then pvulnstatus $cve UNK "Couldn't find kernel image or tools missing to execute the checks" explain "Re-run this script with root privileges, after installing the missing tools indicated above" else pvulnstatus $cve VULN "Kernel source needs to be patched to mitigate the vulnerability" explain "Your kernel is too old to have the mitigation for Variant 1, you should upgrade to a newer kernel. If you're using a Linux distro and didn't compile the kernel yourself, you should upgrade your distro to get a newer kernel." fi else if [ "$msg" = "Vulnerable" ] && [ -n "$v1_mask_nospec" ]; then pvulnstatus $cve OK "Kernel source has been patched to mitigate the vulnerability (silent backport of array_index_mask_nospec)" else if [ "$msg" = "Vulnerable" ]; then msg="Kernel source needs to be patched to mitigate the vulnerability" _explain="Your kernel is too old to have the mitigation for Variant 1, you should upgrade to a newer kernel. If you're using a Linux distro and didn't compile the kernel yourself, you should upgrade your distro to get a newer kernel." fi pvulnstatus $cve "$status" "$msg" [ -n "$_explain" ] && explain "$_explain" unset _explain fi fi } check_variant1_bsd() { cve='CVE-2017-5753' if ! is_cpu_vulnerable 1; then # override status & msg in case CPU is not vulnerable after all pvulnstatus $cve OK "your CPU vendor reported your CPU model as not vulnerable" else pvulnstatus $cve VULN "no mitigation for BSD yet" fi } ################### # SPECTRE VARIANT 2 check_variant2() { _info "\033[1;34mCVE-2017-5715 [branch target injection] aka 'Spectre Variant 2'\033[0m" if [ "$os" = Linux ]; then check_variant2_linux elif echo "$os" | grep -q BSD; then check_variant2_bsd else _warn "Unsupported OS ($os)" fi } check_variant2_linux() { status=UNK sys_interface_available=0 msg='' if sys_interface_check "/sys/devices/system/cpu/vulnerabilities/spectre_v2"; then # this kernel has the /sys interface, trust it over everything sys_interface_available=1 fi if [ "$opt_sysfs_only" != 1 ]; then _info "* Mitigation 1" ibrs_can_tell=0 ibrs_supported='' ibrs_enabled='' ibpb_can_tell=0 ibpb_supported='' ibpb_enabled='' if [ "$opt_live" = 1 ]; then # in live mode, we can check for the ibrs_enabled file in debugfs # all versions of the patches have it (NOT the case of IBPB or KPTI) ibrs_can_tell=1 mount_debugfs for dir in \ /sys/kernel/debug \ /sys/kernel/debug/x86 \ /proc/sys/kernel; do if [ -e "$dir/ibrs_enabled" ]; then # if the file is there, we have IBRS compiled-in # /sys/kernel/debug/ibrs_enabled: vanilla # /sys/kernel/debug/x86/ibrs_enabled: Red Hat (see https://access.redhat.com/articles/3311301) # /proc/sys/kernel/ibrs_enabled: OpenSUSE tumbleweed specex_knob_dir=$dir ibrs_supported="$dir/ibrs_enabled exists" ibrs_enabled=$(cat "$dir/ibrs_enabled" 2>/dev/null) _debug "ibrs: found $dir/ibrs_enabled=$ibrs_enabled" # if ibrs_enabled is there, ibpb_enabled will be in the same dir if [ -e "$dir/ibpb_enabled" ]; then # if the file is there, we have IBPB compiled-in (see note above for IBRS) ibpb_supported="$dir/ibpb_enabled exists" ibpb_enabled=$(cat "$dir/ibpb_enabled" 2>/dev/null) _debug "ibpb: found $dir/ibpb_enabled=$ibpb_enabled" else _debug "ibpb: $dir/ibpb_enabled file doesn't exist" fi break else _debug "ibrs: $dir/ibrs_enabled file doesn't exist" fi done # on some newer kernels, the spec_ctrl_ibrs flag in "$procfs/cpuinfo" # is set when ibrs has been administratively enabled (usually from cmdline) # which in that case means ibrs is supported *and* enabled for kernel & user # as per the ibrs patch series v3 if [ -z "$ibrs_supported" ]; then if grep ^flags "$procfs/cpuinfo" | grep -qw spec_ctrl_ibrs; then _debug "ibrs: found spec_ctrl_ibrs flag in $procfs/cpuinfo" ibrs_supported="spec_ctrl_ibrs flag in $procfs/cpuinfo" # enabled=2 -> kernel & user ibrs_enabled=2 # XXX and what about ibpb ? fi fi if [ -e "/sys/devices/system/cpu/vulnerabilities/spectre_v2" ]; then # when IBPB is enabled on 4.15+, we can see it in sysfs if grep -q ', IBPB' "/sys/devices/system/cpu/vulnerabilities/spectre_v2"; then _debug "ibpb: found enabled in sysfs" [ -z "$ibpb_supported" ] && ibpb_supported='IBPB found enabled in sysfs' [ -z "$ibpb_enabled" ] && ibpb_enabled=1 fi # when IBRS_FW is enabled on 4.15+, we can see it in sysfs if grep -q ', IBRS_FW' "/sys/devices/system/cpu/vulnerabilities/spectre_v2"; then _debug "ibrs: found IBRS_FW in sysfs" [ -z "$ibrs_supported" ] && ibrs_supported='found IBRS_FW in sysfs' ibrs_fw_enabled=1 fi # when IBRS is enabled on 4.15+, we can see it in sysfs if grep -q 'Indirect Branch Restricted Speculation' "/sys/devices/system/cpu/vulnerabilities/spectre_v2"; then _debug "ibrs: found IBRS in sysfs" [ -z "$ibrs_supported" ] && ibrs_supported='found IBRS in sysfs' [ -z "$ibrs_enabled" ] && ibrs_enabled=3 fi fi # in live mode, if ibrs or ibpb is supported and we didn't find these are enabled, then they are not [ -n "$ibrs_supported" ] && [ -z "$ibrs_enabled" ] && ibrs_enabled=0 [ -n "$ibpb_supported" ] && [ -z "$ibpb_enabled" ] && ibpb_enabled=0 fi if [ -z "$ibrs_supported" ]; then check_redhat_canonical_spectre if [ "$redhat_canonical_spectre" = 1 ]; then ibrs_supported="Red Hat/Ubuntu variant" ibpb_supported="Red Hat/Ubuntu variant" fi fi if [ -z "$ibrs_supported" ] && [ -n "$kernel" ]; then if ! which "${opt_arch_prefix}strings" >/dev/null 2>&1; then : else ibrs_can_tell=1 ibrs_supported=$("${opt_arch_prefix}strings" "$kernel" | grep -Fw -e ', IBRS_FW' | head -1) if [ -n "$ibrs_supported" ]; then _debug "ibrs: found ibrs evidence in kernel image ($ibrs_supported)" ibrs_supported="found '$ibrs_supported' in kernel image" fi fi fi if [ -z "$ibrs_supported" ] && [ -n "$opt_map" ]; then ibrs_can_tell=1 if grep -q spec_ctrl "$opt_map"; then ibrs_supported="found spec_ctrl in symbols file" _debug "ibrs: found '*spec_ctrl*' symbol in $opt_map" fi fi # recent (4.15) vanilla kernels have IBPB but not IBRS, and without the debugfs tunables of Red Hat # we can detect it directly in the image if [ -z "$ibpb_supported" ] && [ -n "$kernel" ]; then if ! which "${opt_arch_prefix}strings" >/dev/null 2>&1; then : else ibpb_can_tell=1 ibpb_supported=$("${opt_arch_prefix}strings" "$kernel" | grep -Fw -e 'ibpb' -e ', IBPB' | head -1) if [ -n "$ibpb_supported" ]; then _debug "ibpb: found ibpb evidence in kernel image ($ibpb_supported)" ibpb_supported="found '$ibpb_supported' in kernel image" fi fi fi _info_nol " * Kernel is compiled with IBRS support: " if [ -z "$ibrs_supported" ]; then if [ "$ibrs_can_tell" = 1 ]; then pstatus yellow NO else # if we're in offline mode without System.map, we can't really know pstatus yellow UNKNOWN "in offline mode, we need the kernel image and System.map to be able to tell" fi else if [ "$opt_verbose" -ge 2 ]; then pstatus green YES "$ibrs_supported" else pstatus green YES fi fi _info_nol " * IBRS enabled and active: " if [ "$opt_live" = 1 ]; then if [ "$ibpb_enabled" = 2 ]; then # if ibpb=2, ibrs is forcefully=0 pstatus blue NO "IBPB used instead of IBRS in all kernel entrypoints" else # 0 means disabled # 1 is enabled only for kernel space # 2 is enabled for kernel and user space # 3 is enabled case "$ibrs_enabled" in 0) if [ "$ibrs_fw_enabled" = 1 ]; then pstatus green YES "for firmware code" else pstatus yellow NO if [ -e "$specex_knob_dir/ibrs_enabled" ]; then _verbose " - To enable, \`echo 1 > $specex_knob_dir/ibrs_enabled' as root. If you don't have hardware support, you'll get an error." fi fi ;; 1) if [ "$ibrs_fw_enabled" = 1 ]; then pstatus green YES "for kernel space and firmware code"; else pstatus green YES "for kernel space"; fi;; 2) if [ "$ibrs_fw_enabled" = 1 ]; then pstatus green YES "for kernel, user space, and firmware code" ; else pstatus green YES "for both kernel and user space"; fi;; 3) if [ "$ibrs_fw_enabled" = 1 ]; then pstatus green YES "for kernel and firmware code"; else pstatus green YES; fi;; *) pstatus yellow UNKNOWN;; esac fi else pstatus blue N/A "not testable in offline mode" fi _info_nol " * Kernel is compiled with IBPB support: " if [ -z "$ibpb_supported" ]; then if [ "$ibpb_can_tell" = 1 ]; then pstatus yellow NO else # if we're in offline mode without System.map, we can't really know pstatus yellow UNKNOWN "in offline mode, we need the kernel image to be able to tell" fi else if [ "$opt_verbose" -ge 2 ]; then pstatus green YES "$ibpb_supported" else pstatus green YES fi fi _info_nol " * IBPB enabled and active: " if [ "$opt_live" = 1 ]; then case "$ibpb_enabled" in "") if [ "$ibrs_supported" = 1 ]; then pstatus yellow UNKNOWN else pstatus yellow NO fi ;; 0) pstatus yellow NO _verbose " - To enable, \`echo 1 > $specex_knob_dir/ibpb_enabled' as root. If you don't have hardware support, you'll get an error." ;; 1) pstatus green YES;; 2) pstatus green YES "IBPB used instead of IBRS in all kernel entrypoints";; *) pstatus yellow UNKNOWN;; esac else pstatus blue N/A "not testable in offline mode" fi _info "* Mitigation 2" _info_nol " * Kernel has branch predictor hardening (arm): " if [ -r "$opt_config" ]; then bp_harden_can_tell=1 bp_harden=$(grep -w 'CONFIG_HARDEN_BRANCH_PREDICTOR=y' "$opt_config") if [ -n "$bp_harden" ]; then pstatus green YES _debug "bp_harden: found '$bp_harden' in $opt_config" fi fi if [ -z "$bp_harden" ] && [ -n "$opt_map" ]; then bp_harden_can_tell=1 bp_harden=$(grep -w bp_hardening_data "$opt_map") if [ -n "$bp_harden" ]; then pstatus green YES _debug "bp_harden: found '$bp_harden' in $opt_map" fi fi if [ -z "$bp_harden" ]; then if [ "$bp_harden_can_tell" = 1 ]; then pstatus yellow NO else pstatus yellow UNKNOWN fi fi _info_nol " * Kernel compiled with retpoline option: " # We check the RETPOLINE kernel options if [ -r "$opt_config" ]; then if grep -q '^CONFIG_RETPOLINE=y' "$opt_config"; then pstatus green YES retpoline=1 # shellcheck disable=SC2046 _debug 'retpoline: found '$(grep '^CONFIG_RETPOLINE' "$opt_config")" in $opt_config" else pstatus yellow NO fi else pstatus yellow UNKNOWN "couldn't read your kernel configuration" fi _info_nol " * Kernel compiled with a retpoline-aware compiler: " # Now check if the compiler used to compile the kernel knows how to insert retpolines in generated asm # For gcc, this is -mindirect-branch=thunk-extern (detected by the kernel makefiles) # See gcc commit https://github.com/hjl-tools/gcc/commit/23b517d4a67c02d3ef80b6109218f2aadad7bd79 # In latest retpoline LKML patches, the noretpoline_setup symbol exists only if CONFIG_RETPOLINE is set # *AND* if the compiler is retpoline-compliant, so look for that symbol # # if there is "retpoline" in the file and NOT "minimal", then it's full retpoline # (works for vanilla and Red Hat variants) if [ "$opt_live" = 1 ] && [ -e "/sys/devices/system/cpu/vulnerabilities/spectre_v2" ]; then if grep -qwi retpoline /sys/devices/system/cpu/vulnerabilities/spectre_v2; then if grep -qwi minimal /sys/devices/system/cpu/vulnerabilities/spectre_v2; then pstatus yellow NO "kernel reports minimal retpoline compilation" else retpoline_compiler=1 pstatus green YES "kernel reports full retpoline compilation" fi else if [ "$retpoline" = 1 ]; then pstatus yellow UNKNOWN else pstatus yellow NO fi fi elif [ -n "$opt_map" ]; then # look for the symbol if grep -qw noretpoline_setup "$opt_map"; then retpoline_compiler=1 pstatus green YES "noretpoline_setup symbol found in System.map" else if [ "$retpoline" = 1 ]; then pstatus yellow UNKNOWN else pstatus yellow NO fi fi elif [ -n "$kernel" ]; then # look for the symbol if which "${opt_arch_prefix}nm" >/dev/null 2>&1; then # the proper way: use nm and look for the symbol if "${opt_arch_prefix}nm" "$kernel" 2>/dev/null | grep -qw 'noretpoline_setup'; then retpoline_compiler=1 pstatus green YES "noretpoline_setup found in kernel symbols" else if [ "$retpoline" = 1 ]; then pstatus yellow UNKNOWN else pstatus yellow NO fi fi elif grep -q noretpoline_setup "$kernel"; then # if we don't have nm, nevermind, the symbol name is long enough to not have # any false positive using good old grep directly on the binary retpoline_compiler=1 pstatus green YES "noretpoline_setup found in kernel" else if [ "$retpoline" = 1 ]; then pstatus yellow UNKNOWN else pstatus yellow NO fi fi else if [ "$retpoline" = 1 ]; then pstatus yellow UNKNOWN "couldn't find your kernel image or System.map" else pstatus yellow NO fi fi # only Red Hat has a tunable to disable it on runtime if [ "$opt_live" = 1 ]; then if [ -e "$specex_knob_dir/retp_enabled" ]; then retp_enabled=$(cat "$specex_knob_dir/retp_enabled" 2>/dev/null) _debug "retpoline: found $specex_knob_dir/retp_enabled=$retp_enabled" _info_nol " * Retpoline is enabled: " if [ "$retp_enabled" = 1 ]; then pstatus green YES else pstatus yellow NO _verbose " - To enable, \`echo 1 > $specex_knob_dir/retp_enabled' as root." fi fi fi # only for information, in verbose mode if [ "$opt_verbose" -ge 2 ]; then _info_nol " * Local gcc is retpoline-aware: " if which gcc >/dev/null 2>&1; then if [ -n "$(gcc -mindirect-branch=thunk-extern --version 2>&1 >/dev/null)" ]; then pstatus blue NO else pstatus green YES fi else pstatus blue NO "gcc is not installed" fi fi elif [ "$sys_interface_available" = 0 ]; then # we have no sysfs but were asked to use it only! msg="/sys vulnerability interface use forced, but it's not available!" status=UNK fi cve='CVE-2017-5715' _explain_hypervisor="An updated CPU microcode will have IBRS/IBPB capabilities indicated in the Hardware Check section above. If you're running under an hypervisor (KVM, Xen, VirtualBox, VMware, ...), the hypervisor needs to be up to date to be able to export the new host CPU flags to the guest. You can run this script on the host to check if the host CPU is IBRS/IBPB. If it is, and it doesn't show up in the guest, upgrade the hypervisor." if ! is_cpu_vulnerable 2; then # override status & msg in case CPU is not vulnerable after all pvulnstatus $cve OK "your CPU vendor reported your CPU model as not vulnerable" elif [ -z "$msg" ]; then # if msg is empty, sysfs check didn't fill it, rely on our own test if [ "$retpoline" = 1 ] && [ "$retpoline_compiler" = 1 ] && [ "$retp_enabled" != 0 ]; then pvulnstatus $cve OK "retpoline mitigates the vulnerability" elif [ -n "$bp_harden" ]; then pvulnstatus $cve OK 'branch predictor hardening mitigates the vulnerability for ARM' elif [ "$opt_live" = 1 ]; then if ( [ "$ibrs_enabled" = 1 ] || [ "$ibrs_enabled" = 2 ] ) && [ "$ibpb_enabled" = 1 ]; then pvulnstatus $cve OK "IBRS+IBPB are mitigating the vulnerability" elif ( [ "$ibrs_enabled" = 1 ] || [ "$ibrs_enabled" = 2 ] ) && [ "$ibpb_enabled" = -1 ]; then # IBPB doesn't seem here on this kernel pvulnstatus $cve OK "IBRS is mitigating the vulnerability" elif [ "$ibpb_enabled" = 2 ]; then pvulnstatus $cve OK "Full IBPB is mitigating the vulnerability" elif [ "$ibrs_supported" = 1 ] && [ "$cpuid_spec_ctrl" != 1 ]; then pvulnstatus $cve VULN "Your kernel is compiled with IBRS but your CPU microcode is lacking support to successfully mitigate the vulnerability" explain "The microcode of your CPU needs to be upgraded. This is usually done at boot time by your kernel (the upgrade is not persistent across reboots which is why it's done at each boot). If you're using a distro, make sure you are up to date, as microcode updates are usually shipped alongside with the distro kernel. If an up-to-date CPU microcode is not available for your model, you might want to use a retpoline-enabled kernel instead. Most recent distros have both IBRS-based mitigation (if the microcode is up to date) and automatically fall-back to retpoline-based mitigation otherwise. If yours doesn't, try a more recent one, or compile your own kernel with retpoline and a retpoline-aware compiler (re-run this script with -v to know if your version of gcc is retpoline-aware). $_explain_hypervisor" else pvulnstatus $cve VULN "IBRS hardware + kernel support OR kernel with retpoline are needed to mitigate the vulnerability" explain "Your kernel doesn't have IBRS-based mitigation nor retpoline-based mitigation. If you're using a distro, you might want to upgrade it to get the most recent kernel. Even when the kernel has IBRS-based mitigation compiled-in, the microcode of your CPU needs to be upgraded to be able to use it. This is usually done at boot time by your kernel (the upgrade is not persistent across reboots which is why it's done at each boot). If you're using a distro, make sure you are up to date, as microcode updates are usually shipped alongside with the distro kernel. If an up-to-date CPU microcode is not available for your model, you might want to use a retpoline-enabled kernel instead. Most recent distros have both IBRS-based mitigation (if the microcode is up to date) and automatically fall-back to retpoline-based mitigation otherwise. If yours doesn't, try a more recent one, or compile your own kernel with retpoline and a retpoline-aware compiler (re-run this script with -v to know if your version of gcc is retpoline-aware). $_explain_hypervisor" fi else if [ -n "$ibrs_supported" ]; then pvulnstatus $cve OK "offline mode: IBRS/IBPB will mitigate the vulnerability if enabled at runtime" elif [ "$retpoline" = 1 ]; then pvulnstatus $cve OK "retpoline mitigates the vulnerability" elif [ "$ibrs_can_tell" = 1 ]; then pvulnstatus $cve VULN "IBRS hardware + kernel support OR kernel with retpoline are needed to mitigate the vulnerability" explain "Your kernel doesn't have IBRS-based mitigation nor retpoline-based mitigation. If you're using a distro, you might want to upgrade it to get the most recent kernel. Even when the kernel has IBRS-based mitigation compiled-in, the microcode of your CPU needs to be upgraded to be able to use it. This is usually done at boot time by your kernel (the upgrade is not persistent across reboots which is why it's done at each boot). If you're using a distro, make sure you are up to date, as microcode updates are usually shipped alongside with the distro kernel. If an up-to-date CPU microcode is not available for your model, you might want to use a retpoline-enabled kernel instead. Most recent distros have both IBRS-based mitigation (if the microcode is up to date) and automatically fall-back to retpoline-based mitigation otherwise. If yours doesn't, try a more recent one, or compile your own kernel with retpoline and a retpoline-aware compiler (re-run this script with -v to know if your version of gcc is retpoline-aware). $_explain_hypervisor" else pvulnstatus $cve UNK "offline mode: not enough information" explain "Re-run this script with root privileges, and give it the kernel image (--kernel), the kernel configuration (--config) and the System.map file (--map) corresponding to the kernel you would like to inspect." fi fi else #1 "Vulnerable" #2 "Vulnerable: Minimal generic ASM retpoline" #2 "Vulnerable: Minimal AMD ASM retpoline" # "Mitigation: Full generic retpoline" # "Mitigation: Full AMD retpoline" # $MITIGATION + ", IBPB" # $MITIGATION + ", IBRS_FW" #5 $MITIGATION + " - vulnerable module loaded" # Red Hat only: #2 "Vulnerable: Minimal ASM retpoline", #3 "Vulnerable: Retpoline without IBPB", #4 "Vulnerable: Retpoline on Skylake+", #5 "Vulnerable: Retpoline with unsafe module(s)", # "Mitigation: Full retpoline", # "Mitigation: Full retpoline and IBRS (user space)", # "Mitigation: IBRS (kernel)", # "Mitigation: IBRS (kernel and user space)", # "Mitigation: IBP disabled", if [ "$msg" = "Vulnerable" ]; then # 1 if [ "$ibrs_supported" = 1 ] && [ "$cpuid_spec_ctrl" != 1 ]; then msg="Your kernel is compiled with IBRS but your CPU microcode is lacking support to successfully mitigate the vulnerability" _explain="The microcode of your CPU needs to be upgraded. This is usually done at boot time by your kernel (the upgrade is not persistent across reboots which is why it's done at each boot). If you're using a distro, make sure you are up to date, as microcode updates are usually shipped alongside with the distro kernel. If an up-to-date CPU microcode is not available for your model, you might want to use a retpoline-enabled kernel instead. Most recent distros have both IBRS-based mitigation (if the microcode is up to date) and automatically fall-back to retpoline-based mitigation otherwise. If yours doesn't, try a more recent one, or compile your own kernel with retpoline and a retpoline-aware compiler (re-run this script with -v to know if your version of gcc is retpoline-aware). $_explain_hypervisor" else msg="IBRS hardware + kernel support OR kernel with retpoline are needed to mitigate the vulnerability" _explain="Your kernel doesn't have IBRS-based mitigation nor retpoline-based mitigation. If you're using a distro, you might want to upgrade it to get the most recent kernel. Even when the kernel has IBRS-based mitigation compiled-in, the microcode of your CPU needs to be upgraded to be able to use it. This is usually done at boot time by your kernel (the upgrade is not persistent across reboots which is why it's done at each boot). If you're using a distro, make sure you are up to date, as microcode updates are usually shipped alongside with the distro kernel. If an up-to-date CPU microcode is not available for your model, you might want to use a retpoline-enabled kernel instead. Most recent distros have both IBRS-based mitigation (if the microcode is up to date) and automatically fall-back to retpoline-based mitigation otherwise. If yours doesn't, try a more recent one, or compile your own kernel with retpoline and a retpoline-aware compiler (re-run this script with -v to know if your version of gcc is retpoline-aware). $_explain_hypervisor" fi elif [ "$status" = "VULN" ] && echo "$msg" | grep -q -i -w retpoline && echo "$msg" | grep -q -i -w minimal; then # 2 msg="Retpoline compiled without a retpoline-aware compiler ($msg)" _explain="Your kernel has retpoline-based mitigation enabled (CONFIG_RETPOLINE), but has not been compiled with a retpoline-aware compiler, hence producing an incomplete mitigation. If you're using a distro kernel, you should upgrade it to get the latest kernel available. Otherwise, you should recompile your own kernel with retpoline and a retpoline-aware compiler (re-run this script with -v to know if your version of gcc is retpoline-aware)." elif [ "$status" = "VULN" ] && echo "$msg" | grep -q -w 'Retpoline without IBPB'; then # 3 _explain="Your kernel has retpoline-based mitigation, but to get a complete mitigation, IBPB has to be enabled in addition to retpoline. To use IBPB, the microcode of your CPU needs to be upgraded. This is usually done at boot time by your kernel (the upgrade is not persistent across reboots which is why it's done at each boot). If you're using a distro, make sure you are up to date, as microcode updates are usually shipped alongside with the distro kernel. $_explain_hypervisor" elif [ "$status" = "VULN" ] && echo "$msg" | grep -q -w 'Retpoline on Skylake'; then # 4 _explain="Your kernel has retpoline-based mitigation, but it is not efficient for Skylake-era or more recent Intel CPU models. To get a completely working mitigation, you should use the IBRS-based one instead. Note that the performance degradation of IBRS that can be observed on most CPUs are greatly reduced for Skylake+, so this is a viable alternative. To use IBRS, the microcode of your CPU needs to be upgraded. This is usually done at boot time by your kernel (the upgrade is not persistent across reboots which is why it's done at each boot). If you're using a distro, make sure you are up to date, as microcode updates are usually shipped alongside with the distro kernel. $_explain_hypervisor" elif [ "$status" = "VULN" ] && echo "$msg" | grep -q -w -e 'unsafe module' -e 'vulnerable module'; then # 5 _explain="Your kernel has retpoline-based mitigation, but non-retpoline kernel modules where loaded, and as they share the same memory space than the kernel itself, they could compromise the kernel. Check \`dmesg\` to see which module(s) are concerned." fi pvulnstatus $cve "$status" "$msg" [ -n "$_explain" ] && explain "$_explain" unset _explain fi unset _explain_hypervisor } check_variant2_bsd() { _info "* Mitigation 1" _info_nol " * Kernel supports IBRS: " ibrs_disabled=$(sysctl -n hw.ibrs_disable 2>/dev/null) if [ -z "$ibrs_disabled" ]; then pstatus yellow NO else pstatus green YES fi _info_nol " * IBRS enabled and active: " ibrs_active=$(sysctl -n hw.ibrs_active 2>/dev/null) if [ "$ibrs_active" = 1 ]; then pstatus green YES else pstatus yellow NO fi _info "* Mitigation 2" _info_nol " * Kernel compiled with RETPOLINE: " if [ -n "$kernel_err" ]; then pstatus yellow UNKNOWN "couldn't check ($kernel_err)" else if ! which "${opt_arch_prefix}readelf" >/dev/null 2>&1; then pstatus yellow UNKNOWN "missing '${opt_arch_prefix}readelf' tool, please install it, usually it's in the binutils package" else nb_thunks=$("${opt_arch_prefix}readelf" -s "$kernel" | grep -c -e __llvm_retpoline_ -e __llvm_external_retpoline_ -e __x86_indirect_thunk_) if [ "$nb_thunks" -gt 0 ]; then retpoline=1 pstatus green YES "found $nb_thunks thunk(s)" else pstatus yellow NO fi fi fi cve='CVE-2017-5715' if ! is_cpu_vulnerable 2; then # override status & msg in case CPU is not vulnerable after all pvulnstatus $cve OK "your CPU vendor reported your CPU model as not vulnerable" elif [ "$retpoline" = 1 ]; then pvulnstatus $cve OK "Retpoline mitigates the vulnerability" elif [ "$ibrs_active" = 1 ]; then pvulnstatus $cve OK "IBRS mitigates the vulnerability" elif [ "$ibrs_disabled" = 0 ]; then pvulnstatus $cve VULN "IBRS is supported by your kernel but your CPU microcode lacks support" elif [ "$ibrs_disabled" = 1 ]; then pvulnstatus $cve VULN "IBRS is supported but administratively disabled on your system" else pvulnstatus $cve VULN "IBRS is needed to mitigate the vulnerability but your kernel is missing support" fi } ######################## # MELTDOWN aka VARIANT 3 check_variant3() { _info "\033[1;34mCVE-2017-5754 [rogue data cache load] aka 'Meltdown' aka 'Variant 3'\033[0m" if [ "$os" = Linux ]; then check_variant3_linux elif echo "$os" | grep -q BSD; then check_variant3_bsd else _warn "Unsupported OS ($os)" fi } check_variant3_linux() { status=UNK sys_interface_available=0 msg='' if sys_interface_check "/sys/devices/system/cpu/vulnerabilities/meltdown"; then # this kernel has the /sys interface, trust it over everything sys_interface_available=1 fi if [ "$opt_sysfs_only" != 1 ]; then _info_nol "* Kernel supports Page Table Isolation (PTI): " kpti_support='' kpti_can_tell=0 if [ -n "$opt_config" ]; then kpti_can_tell=1 kpti_support=$(grep -w -e CONFIG_PAGE_TABLE_ISOLATION=y -e CONFIG_KAISER=y -e CONFIG_UNMAP_KERNEL_AT_EL0=y "$opt_config") if [ -n "$kpti_support" ]; then _debug "kpti_support: found option '$kpti_support' in $opt_config" fi fi if [ -z "$kpti_support" ] && [ -n "$opt_map" ]; then # it's not an elif: some backports don't have the PTI config but still include the patch # so we try to find an exported symbol that is part of the PTI patch in System.map # parse_kpti: arm kpti_can_tell=1 kpti_support=$(grep -w -e kpti_force_enabled -e parse_kpti "$opt_map") if [ -n "$kpti_support" ]; then _debug "kpti_support: found '$kpti_support' in $opt_map" fi fi if [ -z "$kpti_support" ] && [ -n "$kernel" ]; then # same as above but in case we don't have System.map and only kernel, look for the # nopti option that is part of the patch (kernel command line option) # 'kpti=': arm kpti_can_tell=1 if ! which "${opt_arch_prefix}strings" >/dev/null 2>&1; then pstatus yellow UNKNOWN "missing '${opt_arch_prefix}strings' tool, please install it, usually it's in the binutils package" else kpti_support=$("${opt_arch_prefix}strings" "$kernel" | grep -w -e nopti -e kpti=) if [ -n "$kpti_support" ]; then _debug "kpti_support: found '$kpti_support' in $kernel" fi fi fi if [ -n "$kpti_support" ]; then if [ "$opt_verbose" -ge 2 ]; then pstatus green YES "found '$kpti_support'" else pstatus green YES fi elif [ "$kpti_can_tell" = 1 ]; then pstatus yellow NO else pstatus yellow UNKNOWN "couldn't read your kernel configuration nor System.map file" fi mount_debugfs _info_nol " * PTI enabled and active: " if [ "$opt_live" = 1 ]; then dmesg_grep="Kernel/User page tables isolation: enabled" dmesg_grep="$dmesg_grep|Kernel page table isolation enabled" dmesg_grep="$dmesg_grep|x86/pti: Unmapping kernel while in userspace" if grep ^flags "$procfs/cpuinfo" | grep -qw pti; then # vanilla PTI patch sets the 'pti' flag in cpuinfo _debug "kpti_enabled: found 'pti' flag in $procfs/cpuinfo" kpti_enabled=1 elif grep ^flags "$procfs/cpuinfo" | grep -qw kaiser; then # kernel line 4.9 sets the 'kaiser' flag in cpuinfo _debug "kpti_enabled: found 'kaiser' flag in $procfs/cpuinfo" kpti_enabled=1 elif [ -e /sys/kernel/debug/x86/pti_enabled ]; then # Red Hat Backport creates a dedicated file, see https://access.redhat.com/articles/3311301 kpti_enabled=$(cat /sys/kernel/debug/x86/pti_enabled 2>/dev/null) _debug "kpti_enabled: file /sys/kernel/debug/x86/pti_enabled exists and says: $kpti_enabled" fi if [ -z "$kpti_enabled" ]; then dmesg_grep "$dmesg_grep"; ret=$? if [ $ret -eq 0 ]; then _debug "kpti_enabled: found hint in dmesg: $dmesg_grepped" kpti_enabled=1 elif [ $ret -eq 2 ]; then _debug "kpti_enabled: dmesg truncated" kpti_enabled=-1 fi fi if [ -z "$kpti_enabled" ]; then _debug "kpti_enabled: couldn't find any hint that PTI is enabled" kpti_enabled=0 fi if [ "$kpti_enabled" = 1 ]; then pstatus green YES elif [ "$kpti_enabled" = -1 ]; then pstatus yellow UNKNOWN "dmesg truncated, please reboot and relaunch this script" else pstatus yellow NO fi else pstatus blue N/A "not testable in offline mode" fi # no security impact but give a hint to the user in verbose mode # about PCID/INVPCID cpuid features that must be present to avoid # too big a performance impact with PTI # refs: # https://marc.info/?t=151532047900001&r=1&w=2 # https://groups.google.com/forum/m/#!topic/mechanical-sympathy/L9mHTbeQLNU _info_nol " * Reduced performance impact of PTI: " grep ^flags "$procfs/cpuinfo" | grep -qw pcid && cpu_pcid=1 grep ^flags "$procfs/cpuinfo" | grep -qw invpcid && cpu_invpcid=1 if [ "$cpu_invpcid" = 1 ]; then pstatus green YES 'CPU supports INVPCID, performance impact of PTI will be greatly reduced' elif [ "$cpu_pcid" = 1 ]; then pstatus green YES 'CPU supports PCID, performance impact o PTI will be reduced' else pstatus blue NO 'PCID and INVPCID not supported, no security impact but performance impact of PTI will be measurable' fi elif [ "$sys_interface_available" = 0 ]; then # we have no sysfs but were asked to use it only! msg="/sys vulnerability interface use forced, but it's not available!" status=UNK fi # Test if the current host is a Xen PV Dom0 / DomU if [ -d "/proc/xen" ]; then # XXX do we have a better way that relying on dmesg? dmesg_grep 'Booting paravirtualized kernel on Xen$'; ret=$? if [ $ret -eq 2 ]; then _warn "dmesg truncated, Xen detection will be unreliable. Please reboot and relaunch this script" elif [ $ret -eq 0 ]; then if [ -e /proc/xen/capabilities ] && grep -q "control_d" /proc/xen/capabilities; then xen_pv_domo=1 else xen_pv_domu=1 fi # PVHVM guests also print 'Booting paravirtualized kernel', so we need this check. dmesg_grep 'Xen HVM callback vector for event delivery is enabled$'; ret=$? if [ $ret -eq 0 ]; then xen_pv_domu=0 fi fi fi if [ "$opt_live" = 1 ]; then # checking whether we're running under Xen PV 64 bits. If yes, we are affected by variant3 # (unless we are a Dom0) _info_nol "* Running as a Xen PV DomU: " if [ "$xen_pv_domu" = 1 ]; then pstatus yellow YES else pstatus blue NO fi fi cve='CVE-2017-5754' if ! is_cpu_vulnerable 3; then # override status & msg in case CPU is not vulnerable after all pvulnstatus $cve OK "your CPU vendor reported your CPU model as not vulnerable" elif [ -z "$msg" ]; then # if msg is empty, sysfs check didn't fill it, rely on our own test if [ "$opt_live" = 1 ]; then if [ "$kpti_enabled" = 1 ]; then pvulnstatus $cve OK "PTI mitigates the vulnerability" elif [ "$xen_pv_domo" = 1 ]; then pvulnstatus $cve OK "Xen Dom0s are safe and do not require PTI" elif [ "$xen_pv_domu" = 1 ]; then pvulnstatus $cve VULN "Xen PV DomUs are vulnerable and need to be run in HVM, PVHVM, PVH mode, or the Xen hypervisor must have the Xen's own PTI patch" explain "Go to https://blog.xenproject.org/2018/01/22/xen-project-spectre-meltdown-faq-jan-22-update/ for more information" elif [ "$kpti_enabled" = -1 ]; then pvulnstatus $cve UNK "couldn't find any clue of PTI activation due to a truncated dmesg, please reboot and relaunch this script" else pvulnstatus $cve VULN "PTI is needed to mitigate the vulnerability" if [ -n "$kpti_support" ]; then if [ -e "/sys/kernel/debug/x86/pti_enabled" ]; then explain "Your kernel supports PTI but it's disabled, you can enable it with \`echo 1 > /sys/kernel/debug/x86/pti_enabled\`" elif grep -q -w nopti -w pti=off /proc/cmdline; then explain "Your kernel supports PTI but it has been disabled on command-line, remove the nopti or pti=off option from your bootloader configuration" else explain "Your kernel supports PTI but it has been disabled, check \`dmesg\` right after boot to find clues why the system disabled it" fi else explain "If you're using a distro kernel, upgrade your distro to get the latest kernel available. Otherwise, recompile the kernel with the CONFIG_PAGE_TABLE_ISOLATION option (named CONFIG_KAISER for some kernels), or the CONFIG_UNMAP_KERNEL_AT_EL0 option (for ARM64)" fi fi else if [ -n "$kpti_support" ]; then pvulnstatus $cve OK "offline mode: PTI will mitigate the vulnerability if enabled at runtime" elif [ "$kpti_can_tell" = 1 ]; then pvulnstatus $cve VULN "PTI is needed to mitigate the vulnerability" explain "If you're using a distro kernel, upgrade your distro to get the latest kernel available. Otherwise, recompile the kernel with the CONFIG_PAGE_TABLE_ISOLATION option (named CONFIG_KAISER for some kernels), or the CONFIG_UNMAP_KERNEL_AT_EL0 option (for ARM64)" else pvulnstatus $cve UNK "offline mode: not enough information" explain "Re-run this script with root privileges, and give it the kernel image (--kernel), the kernel configuration (--config) and the System.map file (--map) corresponding to the kernel you would like to inspect." fi fi else if [ "$xen_pv_domo" = 1 ]; then msg="Xen Dom0s are safe and do not require PTI" status="OK" elif [ "$xen_pv_domu" = 1 ]; then msg="Xen PV DomUs are vulnerable and need to be run in HVM, PVHVM, PVH mode, or the Xen hypervisor must have the Xen's own PTI patch" status="VULN" _explain="Go to https://blog.xenproject.org/2018/01/22/xen-project-spectre-meltdown-faq-jan-22-update/ for more information" elif [ "$msg" = "Vulnerable" ]; then msg="PTI is needed to mitigate the vulnerability" _explain="If you're using a distro kernel, upgrade your distro to get the latest kernel available. Otherwise, recompile the kernel with the CONFIG_PAGE_TABLE_ISOLATION option (named CONFIG_KAISER for some kernels), or the CONFIG_UNMAP_KERNEL_AT_EL0 option (for ARM64)" fi pvulnstatus $cve "$status" "$msg" [ -z "$_explain" ] && [ "$msg" = "Vulnerable" ] && _explain="If you're using a distro kernel, upgrade your distro to get the latest kernel available. Otherwise, recompile the kernel with the CONFIG_PAGE_TABLE_ISOLATION option (named CONFIG_KAISER for some kernels), or the CONFIG_UNMAP_KERNEL_AT_EL0 option (for ARM64)" [ -n "$_explain" ] && explain "$_explain" unset _explain fi # Warn the user about XSA-254 recommended mitigations if [ "$xen_pv_domo" = 1 ]; then _warn _warn "This host is a Xen Dom0. Please make sure that you are running your DomUs" _warn "in HVM, PVHVM or PVH mode to prevent any guest-to-host / host-to-guest attacks." _warn _warn "See https://blog.xenproject.org/2018/01/22/xen-project-spectre-meltdown-faq-jan-22-update/ and XSA-254 for details." fi } check_variant3_bsd() { _info_nol "* Kernel supports Page Table Isolation (PTI): " kpti_enabled=$(sysctl -n vm.pmap.pti 2>/dev/null) if [ -z "$kpti_enabled" ]; then pstatus yellow NO else pstatus green YES fi _info_nol " * PTI enabled and active: " if [ "$kpti_enabled" = 1 ]; then pstatus green YES else pstatus yellow NO fi cve='CVE-2017-5754' if ! is_cpu_vulnerable 3; then # override status & msg in case CPU is not vulnerable after all pvulnstatus $cve OK "your CPU vendor reported your CPU model as not vulnerable" elif [ "$kpti_enabled" = 1 ]; then pvulnstatus $cve OK "PTI mitigates the vulnerability" elif [ -n "$kpti_enabled" ]; then pvulnstatus $cve VULN "PTI is supported but disabled on your system" else pvulnstatus $cve VULN "PTI is needed to mitigate the vulnerability" fi } if [ "$opt_no_hw" = 0 ] && [ -z "$opt_arch_prefix" ]; then check_cpu check_cpu_vulnerabilities _info fi # now run the checks the user asked for if [ "$opt_variant1" = 1 ] || [ "$opt_allvariants" = 1 ]; then check_variant1 _info fi if [ "$opt_variant2" = 1 ] || [ "$opt_allvariants" = 1 ]; then check_variant2 _info fi if [ "$opt_variant3" = 1 ] || [ "$opt_allvariants" = 1 ]; then check_variant3 _info fi _vars=$(set | grep -Ev '^[A-Z_[:space:]]' | sort | tr "\n" '|') _debug "variables at end of script: $_vars" _info "A false sense of security is worse than no security at all, see --disclaimer" if [ "$opt_batch" = 1 ] && [ "$opt_batch_format" = "nrpe" ]; then if [ ! -z "$nrpe_vuln" ]; then echo "Vulnerable:$nrpe_vuln" else echo "OK" fi fi if [ "$opt_batch" = 1 ] && [ "$opt_batch_format" = "json" ]; then _echo 0 "${json_output%?}]" fi if [ "$opt_batch" = 1 ] && [ "$opt_batch_format" = "prometheus" ]; then echo "# TYPE specex_vuln_status untyped" echo "# HELP specex_vuln_status Exposure of system to speculative execution vulnerabilities" echo "$prometheus_output" fi # exit with the proper exit code [ "$global_critical" = 1 ] && exit 2 # critical [ "$global_unknown" = 1 ] && exit 3 # unknown exit 0 # ok