minix/releasetools/x86_hdimage.sh
David van Moolenbroek e4d99eb9b0 Basic live rerandomization infrastructure
This commits adds a basic infrastructure to support Address Space
Randomization (ASR).  In a nutshell, using the already imported ASR
LLVM pass, multiple versions can be generated for the same system
service, each with a randomized, different address space layout.
Combined with the magic instrumentation for state transfer, a system
service can be live updated into another ASR-randomized version at
runtime, thus providing live rerandomization.

Since MINIX3 is not yet capable of running LLVM linker passes, the
ASR-randomized service binaries have to be pregenerated during
crosscompilation.  These pregenerated binaries can then be cycled
through at runtime.  This patch provides the basic proof-of-concept
infrastructure for both these parts.

In order to support pregeneration, the clientctl host script has
been extended with a "buildasr" command.  It is to be used after
building the entire system with bitcode and magic support, and will
produce a given number of ASR-randomized versions of all system
services.  These services are placed in /usr/service/asr in the
image that is generated as final step by the "buildasr" command.

In order to support runtime updating, a new update_asr(8) command
has been added to MINIX3.  This command attempts to live-update the
running system services into their next ASR-randomized versions.
For now, this command is not run automatically, and thus must be
invoked manually.

Technical notes:

- For various reasons, magic instrumentation is x86-only for now,
  and ASR functionality is therefore to be used on x86 only as well.
- The ASR-randomized binaries are placed in numbered subdirectories
  so as not to have to change their actual program names, which are
  assumed to be static in various places (system.conf, procfs).
- The root partition is typically too small to contain all the
  produced binaries, which is why we introduce /usr/service.  There
  is a symlink from /service/asr to /usr/service/asr for no other
  reason than to let userland continue to assume that all services
  are reachable through /service.
- The ASR count field (r_asr_count/ASRcount) maintained by RS is not
  used within RS in any way; it is only passed through procfs to
  userland in order to allow update_asr(8) to keep track of which
  version is currently loaded without having to maintain own state.
- Ideally, pre-instrumentation linking of a service would remove all
  its randomized versions.  Currently, the user is assumed not to
  perform ASR instrumentation and then recompile system services
  without performing ASR instrumentation again, as the randomized
  binaries included in the image would then be stale.  This aspect
  has to be improved later.
- Various other issues are flagged in the comments of the various
  parts of this patch.

Change-Id: I093ad57f31c18305591f64b2d491272288aa0937
2015-09-17 17:15:03 +00:00

220 lines
6.6 KiB
Bash
Executable file

#!/usr/bin/env bash
set -e
#
# This script creates a bootable image and should at some point in the future
# be replaced by makefs.
#
# Supported command line switches:
# -i build iso image instead of qemu imaeg
# -b bitcode build, increase partition sizes as necessary
#
: ${ARCH=i386}
: ${OBJ=../obj.${ARCH}}
: ${CROSS_TOOLS=${OBJ}/"tooldir.`uname -s`-`uname -r`-`uname -m`"/bin}
: ${CROSS_PREFIX=${CROSS_TOOLS}/i586-elf32-minix-}
: ${JOBS=1}
: ${DESTDIR=${OBJ}/destdir.$ARCH}
: ${RELEASETOOLSDIR=./releasetools/}
: ${FSTAB=${DESTDIR}/etc/fstab}
: ${BUILDVARS=}
: ${BUILDSH=build.sh}
: ${CREATE_IMAGE_ONLY=0}
: ${RC=minix_x86.rc}
#
# Directory where to store temporary file system images
#
: ${IMG_DIR=${OBJ}/img}
: ${CDFILES=${IMG_DIR}/cd}
# All sized are written in 512 byte blocks
#
# we create a disk image of about 2 gig's
# for alignment reasons, prefer sizes which are multiples of 4096 bytes
#
# these sizes are insufficient for bitcode builds!
# invoke this script with the -b flag to increase sizes accordingly
#
: ${ROOT_SIZE=$(( 64*(2**20) / 512))}
: ${HOME_SIZE=$(( 128*(2**20) / 512))}
: ${USR_SIZE=$(( 1792*(2**20) / 512))}
#
# Do some math to determine the start addresses of the partitions.
# Don't leave holes so the 'partition' invocation later is easy.
#
# Where the kernel & boot modules will be
MODDIR=${DESTDIR}/boot/minix/.temp
while getopts "ib" c
do
case "$c" in
i) : ${IMG=minix_x86.iso}
ISOMODE=1
;;
b) # bitcode build: increase partition sizes
ROOT_SIZE="$((${ROOT_SIZE} + 192*(2**20) / 512))"
USR_SIZE="$((${USR_SIZE} + 256*(2**20) / 512))"
;;
esac
done
: ${IMG=minix_x86.img}
if [ "x${ISOMODE}" = "x1" ]
then
# In iso mode, make all FSes fit (i.e. as small as possible), but
# leave some space on /
ROOTSIZEARG="-x 5"
else
# In hd image mode, FSes have fixed sizes
ROOTSIZEARG="-b $((${ROOT_SIZE} / 8))"
USRSIZEARG="-b $((${USR_SIZE} / 8))"
HOMESIZEARG="-b $((${HOME_SIZE} / 8))"
fi
if [ ! -f ${BUILDSH} ]
then
echo "Please invoke me from the root source dir, where ${BUILDSH} is."
exit 1
fi
export PATH=/bin:/sbin:/usr/bin:/usr/sbin:/usr/local/bin:/usr/local/sbin:${PATH}
#
# Are we going to build the minix sources?
#
if [ ${CREATE_IMAGE_ONLY} -eq 1 ]
then
if [ ! -d ${DESTDIR} ]
then
echo "Minix source code does'nt appear to have been built."
echo "Please try with \$CREATE_IMAGE_ONLY set to 0."
exit 1
fi
fi
#
# Artifacts from this script are stored in the IMG_DIR
#
rm -rf ${IMG_DIR} ${IMG}
mkdir -p ${IMG_DIR} ${CDFILES}
if [ ${CREATE_IMAGE_ONLY} -eq 0 ]
then
echo "Going to build Minix source code..."
#
# Remove the generated files to allow us call build.sh without '-V SLOPPY_FLIST=yes'.
#
rm -f ${FSTAB}
#
# Now start the build.
#
sh ${BUILDSH} -j ${JOBS} -m ${ARCH} -O ${OBJ} -D ${DESTDIR} ${BUILDVARS} -U -u distribution
else
${CROSS_TOOLS}/nbmake-i386 -C releasetools do-hdboot
fi
#
# create a fstab entry in /etc this is normally done during the
# setup phase on x86
#
cat >${FSTAB} <<END_FSTAB
/dev/c0d0p2 /usr mfs rw 0 2
/dev/c0d0p3 /home mfs rw 0 2
none /sys devman rw,rslabel=devman 0 0
none /dev/pts ptyfs rw,rslabel=ptyfs 0 0
END_FSTAB
rm -f ${DESTDIR}/SETS.*
${CROSS_TOOLS}/nbpwd_mkdb -V 0 -p -d ${DESTDIR} ${DESTDIR}/etc/master.passwd
#
# make the different file system. this part is *also* hacky. We first convert
# the METALOG.sanitised using mtree into a input METALOG containing uids and
# gids.
# After that we do some magic processing to add device nodes (also missing from METALOG)
# and convert the METALOG into a proto file that can be used by mkfs.mfs
#
echo "Creating the file systems"
#
# read METALOG and use mtree to convert the user and group names into uid and gids
# FIX put "input somewhere clean"
#
cat ${DESTDIR}/METALOG.sanitised | ${CROSS_TOOLS}/nbmtree -N ${DESTDIR}/etc -C -K device > ${IMG_DIR}/input
# add rc (if any)
if [ -f ${RC} ]; then
cp ${RC} ${DESTDIR}/usr/etc/rc.local
echo "./usr/etc/rc.local type=file uid=0 gid=0 mode=0644" >> ${IMG_DIR}/input
fi
# add fstab
echo "./etc/fstab type=file uid=0 gid=0 mode=0755 size=747 time=1365060731.000000000" >> ${IMG_DIR}/input
# add any generated ASR-randomized service binaries (but not their root directory, which is already there)
# TODO: apply stricter file permissions for both these and the base /service binaries, against local attacks
(cd ${DESTDIR} && find ./usr/service/asr -type d | sed '1d;s/$/ type=dir uid=0 gid=0 mode=0755/') >> ${IMG_DIR}/input
(cd ${DESTDIR} && find ./usr/service/asr -type f | sed 's/$/ type=file uid=0 gid=0 mode=0755/') >> ${IMG_DIR}/input
# fill root.img (skipping /usr entries while keeping the /usr directory)
cat ${IMG_DIR}/input | grep -v "^./usr/" | ${CROSS_TOOLS}/nbtoproto -b ${DESTDIR} -o ${IMG_DIR}/root.proto
#
# Create proto files for /usr and /home using toproto.
#
cat ${IMG_DIR}/input | grep "^\./usr/\|^. " | sed "s,\./usr,\.,g" | ${CROSS_TOOLS}/nbtoproto -b ${DESTDIR}/usr -o ${IMG_DIR}/usr.proto
cat ${IMG_DIR}/input | grep "^\./home/\|^. " | sed "s,\./home,\.,g" | ${CROSS_TOOLS}/nbtoproto -b ${DESTDIR}/home -o ${IMG_DIR}/home.proto
if [ "x${ISOMODE}" = "x1" ]
then
cp ${DESTDIR}/usr/mdec/boot_monitor ${CDFILES}/boot
cp ${MODDIR}/* ${CDFILES}/
. ${RELEASETOOLSDIR}/release.functions
cd_root_changes # uses $CDFILES and writes $CDFILES/boot.cfg
# start the image off with the iso image; reduce root size to reserve
${CROSS_TOOLS}/nbwriteisofs -s0x0 -l MINIX -B ${DESTDIR}/usr/mdec/bootxx_cd9660 -n ${CDFILES} ${IMG}
ISO_SIZE=$((`${CROSS_TOOLS}/nbstat -f %z ${IMG}` / 512))
else
# just make an empty iso partition
ISO_SIZE=8
fi
#
# Generate /root, /usr and /home partition images.
#
echo "Writing Minix filesystem images"
ROOT_START=${ISO_SIZE}
echo " - ROOT"
_ROOT_SIZE=$((`${CROSS_TOOLS}/nbmkfs.mfs -d ${ROOTSIZEARG} -I $((${ROOT_START}*512)) ${IMG} ${IMG_DIR}/root.proto`/512))
USR_START=$((${ROOT_START} + ${_ROOT_SIZE}))
echo " - USR"
_USR_SIZE=$((`${CROSS_TOOLS}/nbmkfs.mfs -d ${USRSIZEARG} -I $((${USR_START}*512)) ${IMG} ${IMG_DIR}/usr.proto`/512))
HOME_START=$((${USR_START} + ${_USR_SIZE}))
echo " - HOME"
_HOME_SIZE=$((`${CROSS_TOOLS}/nbmkfs.mfs -d ${HOMESIZEARG} -I $((${HOME_START}*512)) ${IMG} ${IMG_DIR}/home.proto`/512))
#
# Write the partition table using the natively compiled
# minix partition utility
#
${CROSS_TOOLS}/nbpartition -m ${IMG} 0 81:${ISO_SIZE} \
81:${_ROOT_SIZE} 81:${_USR_SIZE} 81:${_HOME_SIZE}
mods="`( cd ${MODDIR}; echo mod* | tr ' ' ',' )`"
if [ "x${ISOMODE}" = "x1" ]
then
echo "CD image at `pwd`/${IMG}"
else
echo "To boot this image on kvm:"
echo "cd ${MODDIR} && qemu-system-i386 --enable-kvm -m 256 -kernel kernel -append \"rootdevname=c0d0p1\" -initrd \"${mods}\" -hda `pwd`/${IMG}"
fi