diff options
author | Zach van Rijn <me@zv.io> | 2022-12-10 11:32:48 -0600 |
---|---|---|
committer | Zach van Rijn <me@zv.io> | 2022-12-10 13:45:42 -0600 |
commit | e6db002716c0c46fd73ebc258ada23edbf5c7d12 (patch) | |
tree | d991697f179e07eaab1c8fdc2dad8f18efbe0111 | |
download | bootstrap-e6db002716c0c46fd73ebc258ada23edbf5c7d12.tar.gz bootstrap-e6db002716c0c46fd73ebc258ada23edbf5c7d12.tar.bz2 bootstrap-e6db002716c0c46fd73ebc258ada23edbf5c7d12.tar.xz bootstrap-e6db002716c0c46fd73ebc258ada23edbf5c7d12.zip |
Initial proof of concept.
Currently, if you run this on a Linux box as an unprivileged user:
$ ./bootstrap armv7 /some/scratch/directory
you will end up with, completely from source:
rootfs-armv7l-adelie-linux-musleabihf-preimage.tgz
which is an armv7-native rootfs with all the tools that you need
to build our distribution. (Substitute your favorite target).
Work that remains to be done involves fixing "bootstrap.sh" in the
packages.git repository, then plumbing that back into this project
so that the complete process can be automated.
-rw-r--r-- | binsub.c | 277 | ||||
-rwxr-xr-x | bootstrap | 763 | ||||
-rwxr-xr-x | setup-abuild | 362 | ||||
-rwxr-xr-x | setup-packages | 13 | ||||
-rwxr-xr-x | setup-preimage | 35 |
5 files changed, 1450 insertions, 0 deletions
diff --git a/binsub.c b/binsub.c new file mode 100644 index 0000000..c66148c --- /dev/null +++ b/binsub.c @@ -0,0 +1,277 @@ +/** + * binsub.c / 2022-12-09 + * + * (C) 2022 Zach van Rijn <me@zv.io> + * + * MIT License + * + * This utility truncates or replaces needles in an input file; + * truncation meaning the replacement string is empty. + * + * Replacement string length must be less than or equal to that + * of the needle because the file length must remain unchanged. + * + * For efficient operation, consider deploying this on a '.tar' + * file instead of individual files within a directory. + */ + +#include <stdio.h> +#include <string.h> +#include <stdlib.h> + +/** + * Basic memory structure. + */ +struct buffer +{ + char *data; + size_t len; +}; + + +/** + * Given a needle ('find') and optional replacement ('repl'), if + * the needle is found, truncate it, inject the replacement, and + * pad the tail end of the matching string with null bytes. + * + * The file length remains the same, and we hope that nobody is + * relying on precomputed offsets into the strings. Mega kludge! + * + * Replacement = "": + * + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * in |.|.|.|N|E|E|D|L|E|.|.|.|O|T|H|E|R| |D|A|T|A|.|.|.|0| + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * ^ shift data ^ + * +-------------------------------+ + * + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * out |.|.|.|.|.|.|O|T|H|E|R| |D|A|T|A|.|.|.|0|0|0|0|0|0|0| + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * ^ shift data ^ + * +-------------------------------+ + * + * Replacement = "FOO": + * + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * in |.|.|.|N|E|E|D|L|E|.|.|.|O|T|H|E|R| |D|A|T|A|.|.|.|0| + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * ^ shift data ^ + * +-------------------------------+ + * + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * out |.|.|.|F|O|O|.|.|.|O|T|H|E|R| |D|A|T|A|.|.|.|0|0|0|0| + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * ^ FOO ^ shift data ^ + * +=====+-------------------------------+ + */ +void +replace (struct buffer *buf, const char *find, const char *repl) +{ + char *match; /* pointer to found needle */ + size_t idx; /* cursor into file buffer */ + + size_t nlen; /* length of needle */ + size_t rlen = 0; /* length of replacement */ + + size_t mlen; /* length of matching string */ + + nlen = strlen(find); + + /** + * Iterate over each character in the current string in the + * buffer, because multiple full matches may be possible. If + * we cannot find a match in the current string, skip to the + * next string. I don't think there is a more optimal way? + */ + for (idx = 0; idx <= buf->len; idx++) + { + /** + * Does the current string contain the needle? + */ + match = strstr(buf->data + idx, find); + if (match) + { + /** + * How long is the current string? We need to search + * it until we cannot find any more matches. + */ + mlen = strlen(match); + printf("%10zu bytes at offset 0x%010lx (%02ld%%)\n", + mlen, + (match - buf->data), + (100 * (match - buf->data)) / buf->len + ); + + /** + * The replacement string is l.e. the length of the + * needle, so if it is non-empty, inject it first. + */ + if (repl) + { + rlen = strlen(repl); + memcpy(match, repl, rlen); + } + + /** + * The replacement length may be zero (if empty). In + * any case, copy the non-needle string remainder to + * the current matched (needle) location plus offset + * of any injected replacement. Zero out the tail. + */ + memmove(match + rlen, match + nlen, mlen - nlen); + memset(match + mlen - nlen + rlen, 0, nlen - rlen); + } + else + { + /** + * This is a partial optimization. Don't bother with + * searching for needles in the rest of this string; + * we already know none exist. + */ + idx += strlen(buf->data + idx); + } + } +} + + +/** + * Read the contents of a file into a newly allocated buffer. It + * is possible to 'mmap()', but it leaves less room for checks. + */ +void +scanner (const char *file, const char *find, const char *repl) +{ + FILE *fp = NULL; + size_t nb = 0; + + struct buffer buf; + memset(&buf, 0, sizeof(struct buffer)); + + fp = fopen(file, "rb+"); + if (!fp) + { + fprintf(stderr, + "E: Could not open FILE '%s' for reading!\n", + file + ); + return; + } + + fseek(fp, 0, SEEK_END); + buf.len = ftell(fp); + fseek(fp, 0, SEEK_SET); + + /** + * Allocate memory for the entire file at once. This is not + * ideal, but we don't expect large files for our use case. + */ + buf.data = malloc(buf.len + 1); + if (!buf.data) + { + fclose(fp); + fprintf(stderr, + "E: Could not allocate '%zu' bytes for file '%s'\n", + buf.len, + file + ); + return; + } + buf.data[buf.len] = 0; /* extra byte needs to be nil */ + + nb = fread(buf.data, 1, buf.len, fp); + if (nb != buf.len) + { + free(buf.data); + buf.data = NULL; + fclose(fp); + fprintf(stderr, + "E: Only read '%zu' / '%zu' bytes of file '%s'\n", + nb, + buf.len, + file + ); + return; + } + fclose(fp); + + printf("Examining file '%s'...\n", file); + replace(&buf, find, repl); + + fp = fopen(file, "wb"); + if (!fp) + { + fprintf(stderr, + "E: Could not open FILE '%s' for writing!\n", + file + ); + return; + } + + nb = fwrite(buf.data, 1, buf.len, fp); + if (nb != buf.len) + { + free(buf.data); + buf.data = NULL; + fclose(fp); + fprintf(stderr, + "E: Only wrote '%zu' / '%zu' bytes of file '%s'\n", + nb, + buf.len, + file + ); + return; + } + fclose(fp); + + free(buf.data); +} + + +/** + * WARNING! + * + * This program replaces all occurrences of NEEDLE within string + * sections of an input file with the string REPLACE. The input + * file is overwritten. Few, if any, sanity checks are in place. + */ +int +main (int argc, char **argv) +{ + char *prog = NULL; + char *file = NULL; + char *find = NULL; + char *repl = NULL; + + prog = argv[0]; + + switch (argc) + { + case 3: + file = argv[1]; + find = argv[2]; + break; + case 4: + file = argv[1]; + find = argv[2]; + repl = argv[3]; + if (strlen(repl) > strlen(find)) + { + fprintf(stderr, + "E: REPLACE cannot be longer than NEEDLE\n" + ); + return 1; + } + break; + default: + fprintf(stderr, + "Usage: %s FILE NEEDLE [REPLACE]\n", + prog + ); + return 1; + } + + scanner(file, find, repl); + + return 0; +} diff --git a/bootstrap b/bootstrap new file mode 100755 index 0000000..d2fcf08 --- /dev/null +++ b/bootstrap @@ -0,0 +1,763 @@ +#!/bin/sh -e + +#=============================================================== +# Filename : bootstrap +# Purpose : Bootstraps Adélie from source for any architecture. +# Authors : Zach van Rijn <me@zv.io> +# License : MIT +# Revision : 20221209 +#=============================================================== + +#=============================================================== +# README +#=============================================================== +# +# overview +# -------- +# +# Given a basic development environment ("Baseline System") that +# contains standard system utilities, this script bootstraps the +# Adélie Linux distribution for any suported target architecture +# (assuming that musl, gcc, etc. have been ported to it) WITHOUT +# requiring 'root' privileges. This process is slow* and costly, +# but demonstrates that bootstrapping from source is possible. +# +# Effort has been focused on correctness instead of performance. +# This means, some files can be deleted sooner and some of the +# binaries can run natively on the build machine. We know and do +# appreciate there are different opinions on how to approach the +# bootstrap process. This meets our needs and hopefully offers a +# different perspective on how to simplify cross compilation. +# +# (*) See the "requirements" section for mitigations/discussion. +# +# +# features +# -------- +# +# * One-click bootstrap. Just start the script and walk away. +# +# $ ./bootstrap.sh ppc64 /path/to/new/scratch/dir +# +# * Can be done without root privileges: no 'chroot(8)'. +# +# * This is the real deal. No "seed" binaries are required to +# go from start to finish. The final output is what we use +# to set up our official build servers totally* from source. +# +# * Can be done on a wide range of platforms, almost certainly +# those supported by mcmtools (want to contribute???). +# +# * Minimal dependencies. Nothing fancy. Shell scripts only. +# +# * Can be adapted to assist with porting Adélie to new target +# architectures. TODO: write a tool to automatically do so. +# +# (*) We don't count binaries in your starting environment, and +# recommend https://www.gnu.org/software/mes/ for the brave. +# +# +# design +# ------ +# +# The goal of this script is to support as wide a range of uses +# as possible. For example, a full cross-platform bootstrap that +# needs to be done without target hardware or virtual machines. +# +# There are a few different strategies, depending on your needs +# and what type of privileges/hardware you have access to. +# +# In this diagram, indentation refers to an output product, and +# moving down vertically refers to an input step. CPU A is the +# build architecture, and CPU B is the target architecture. Your +# mileage may vary, and you can mix/match other approaches.. +# +# * CPU A --> CPU B (slowest: no privilege, no CPU B hardware) +# --> Bootstrap 1 (Cross Libc + Partial Cross CPU B) +# --> CPU A + CPU B Mixed Rootfs +# --> Partial Emulation (PRoot) +# --> Bootstrap 2 (Full Cross CPU B) +# --> CPU B Host Rootfs +# --> Full Emulation (PRoot) +# --> Adélie Bootstrap +# --> Adélie Rootfs for CPU B +# +# * CPU A --> CPU B (slower: privilege, no CPU B hardware) +# --> Bootstrap 1 (Cross Libc + Partial Cross CPU B) +# --> CPU A + CPU B Mixed Rootfs +# --> Partial Emulation (registered binfmt_misc) +# --> Bootstrap 2 (Full Cross CPU B) +# --> CPU B Host Rootfs +# --> Full Emulation (registered binfmt_misc) +# --> Adélie Bootstrap +# --> Adélie Rootfs for CPU B +# +# * CPU A --> CPU B (faster; no privilege, CPU B hardware) +# --> Bootstrap 1 (Cross Libc + Partial Cross CPU B) +# --> CPU A + CPU B Mixed Rootfs +# --> Partial Emulation (PRoot) +# --> Bootstrap 2 (Full Cross CPU B) +# --> CPU B Host Rootfs +# --> Copy to Native CPU B Hardware +# --> Adélie Bootstrap +# --> Adélie Rootfs for CPU B +# +# * CPU B --> CPU B (fastest; QEMU-system or native hardware) +# --> Bootstrap 1 (Cross Libc) +# --> CPU B Mixed Rootfs +# --> Bootstrap 2 (Remove Contamination) +# --> CPU B Host Rootfs +# --> Adélie Bootstrap +# --> Adélie Rootfs for CPU B +# +# The resulting Adélie Rootfs is capable of building the entire +# Adélie Linux distribution from source. We use it to seed our +# official build servers. This removes the dependency on using +# previous binary releases of our distribution, which had been +# manually bootstrapped way back from Gentoo and/or on a G5. +# +# +# requirements +# ------------ +# +# TL;DR: You must be able to run the mcmtools bootstrap script. +# +# Cross builds take up a lot of TIME and SPACE, and this cannot +# be improved by throwing hardware at it. This is primarily due +# to the tradeoff of not requiring root privileges at any point, +# and the decision to emulate a native environment instead of to +# force explicit cross-compilation at each step in the process. +# +# Targeting an architecture of opposite endianness will be even +# slower; this is a QEMU limitation. +# +# If you wish to bootstrap to the same target CPU architecture, +# or a compatible mode (e.g. i686 on x86_64), it will be faster. +# +# mcmtools is a hard dependency for our bootstrap process now: +# +# https://git.zv.io/toolchains/bootstrap +# +# It is a simple analog to Buildroot or Yocto. Those tools could +# be used, too, and would provide the "host rootfs" environment. +# +# Internet access is required, but (as an exercise left to the +# reader) it is possible to pre-download all required sources if +# you provide your own rootfs for the static QEMU build process. +# +# If you are in a position to use native hardware, then you are +# able to get away with only the final stages of bootstrapping. +# To do this, you'd essentially comment out the first stages or +# copy the results of the first stages elsewhere and continue. +# +# Other requirements that you should be aware of, estimated: +# +# * As many CPU cores as you can throw at it; +# +# * ~15 GB for each mcmtools rootfs ("seed" and "host"); +# +# * ~ 3 GB for toolchains; +# +# * ~ 2 GB for each intermediate rootfs; +# +# * ~ 2 GB for the "system/" package repository, when built; +# +# * Please refer to the README in 'bootstrap', linked above, +# for more information about performance. Most of this can +# be gained back if you adapt this script to use 'chroot(8)' +# instead of 'PRoot', at the expense of requiring privilege, +# with the correct registration of QEMU with 'binfmt_misc'. +# +# In brief, there is an approximate factor of 13 slowdown on +# practical workloads when using QEMU user and 'PRoot' to do +# dynamic binary translation and emulate a foreign machine. +# +# +# process +# ------- +# +# The illustration below outlines the complete bootstrap process +# and roughly corresponds to the script layout/organization. Our +# terminology is not perfectly consistent; please excuse this. +# +# The term "build" is shorthand for "build-native" CPU, which +# is the machine on which you are performing the bootstrap. +# +# The term "native" is shorthand for "foreign-native" CPU, which +# is the machine to which you are targeting the bootstrap, and +# relates to the "host" CPU on which the code will run, but that +# with the help of dynamic binary translation, runs "natively". +# +# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - +# build (unstable, everything provided by user) +# +# +-----------------+ User-provided tools. Dependency +# | Baseline System | of 'mcmtools', which will verify +# +-----------------+ that these tools are available. +# | +# +----------+ Script to build pinned versions +# | mcmtools | of common system utilities, a +# +----------+ host-arch host-libc toolchain, a +# | host-arch musl-libc toolchain... +# | +# - - - - -|- - - - - - - - - - - - - - - - - - - - - - - - +# mixed | (stable versions, unstable libc) +# | +# +-------------+ ...and a 'chroot'-able rootfs. A +# | seed rootfs | sane, but not clean, environment +# +-------------+ in which we begin the bootstrap. +# | +# +-----------+ Script to build 'PRoot' and its +# | emulators | dependencies, as well as static +# +-----------+ QEMU user binaries. Add to seed. +# | +# +-----------------+ Static musl toolchains targeting +# | musl toolchains | a given architecture: cross from +# +-----------------+ the host CPU & foreign "native". +# | Output binaries will run via the +# | 'binfmt_misc' mechanism + QEMU. +# | +# +-------------+ Script to build Alpine Package +# | build tools | Keeper (APK) and dependencies. +# +-------------+ All binaries are cross-compiled! +# | This step is a sanity check. +# | +# +----------+ Script to build pinned versions +# | mcmtools | of common system utilities. This +# +----------+ time, all utilities are target- +# | native. Build more toolchains... +# | +# - - - - -|- - - - - - - - - - - - - - - - - - - - - - - - +# host | (stable versions, stable libc) +# | +# +-------------+ ...and a 'chroot'-able rootfs. A +# | host rootfs | sane, clean, foreign "native" +# +-------------+ rootfs that requires QEMU, or is +# | able to run on native hardware. +# | +# +-------------+ Script to build Alpine Package +# | build tools | Keeper (APK) and dependencies. +# +-------------+ These binaries are native built! +# | This step is required! +# | +# +-------------+ Script to build the Adélie Linux +# | system repo | "system/" package repository. It +# +-------------+ is used to build core packages. +# | +# +---------------+ Script to install packages into +# | image creator | a clean rootfs. Carryover from +# +---------------+ the mcmtools process is removed. +# | +# - - - - -|- - - - - - - - - - - - - - - - - - - - - - - - +# verify | (optional verification) +# | +# ... The "host" stage can be repeated +# | zero or more times to ensure the +# | final image is not contaminated. +# | +# - - - - -|- - - - - - - - - - - - - - - - - - - - - - - - +# output | (Welcome to Adélie Linux) +# | +# +---------------+ This is a minimal Adélie rootfs. +# | adelie rootfs | Copy it to native hardware, then +# +---------------+ use it as a builder 'chroot'. It +# is used to seed official Adélie +# build boxen (e.g. autobuilder). +# +# limitations +# ----------- +# +# The builds are not hermetically sealed. That is not the point. +# The build environment should not *need* anything from outside, +# but it is not *prevented* from accessing anything. You should +# run this on a clean, trusted machine. +# +# There is no guarantee of byte-for-byte reproducible builds at +# this time. This is, in part, due to timestamps and tar headers +# but may involve a lack of determinism in parallel builds. +# +# QEMU user emulation may cause subtle incompatibilites with the +# target CPU hardware. For example, CPU feature (mis)detection. +# +# It is not currently possible to safely reuse any of the output +# if the target architecture is changed. You will need to start +# from scratch if you wish to change the target. This is a TODO. +# +# +# todo +# ---- +# +# * Do not inherit the build environment; avoid contamination. +# +# * Finish the "system/" build; the "bootstrap.sh" script will +# require some work. Current status: preimage is finished. +# + +HERE="$(dirname $(readlink -f ${0}))"; + + +#--------------------------------------------------------------- +# initialization + +## +# Haaaalp! +# +usage () +{ + printf "Usage: %s ARCH BASE\n\n" "${0}"; + cat <<EOF + ARCH { aarch64, armv7, ppc64, ppc, x86_64, pmmx } + + BASE an absolute path to bootstrap out of. + ** MAY CAUSE DATA LOSS IF SET INCORRECTLY! ** +EOF + exit 0; +} + + +## +# argv[1]: ARCH +# +# ARCH is the Adélie Linux target. This is the first step in the +# porting process, so e.g. mips64, sparc64, riscv64 will need to +# be added to this table when the time comes to port to them. +# +# ARCH is translated to canonical GCC and QEMU machine types. +# +case "${1}" in +# adelie gcc qemu +# ------ --- ---- + aarch64) m=aarch64: ; q=aarch64 ; ;; + armv7) m=armv7l:eabihf ; q=arm ; ;; + ppc) m=powerpc: ; q=ppc ; ;; + ppc64) m=powerpc64: ; q=ppc64 ; ;; + riscv64) m=riscv64: ; q=riscv64 ; ;; + x86_64) m=x86_64: ; q=x86_64 ; ;; + pmmx) m=i586: ; q=i386 ; ;; + *) usage ;; +esac +shift; + + +## +# argv[2]: BASE +# +# BASE is a semi-permanent scratch directory. It is where all of +# the magic happens, and probably cannot be relocated easily. Be +# careful to not set it incorrectly or to a place you'd regret +# being overwritten, corrupted, or deleted. +# +case "${1}" in + /*) BASE="${1}"; + ;; + *) printf "BASE not set, or not an absolute path!\n"; + exit 1; + ;; +esac +shift; + + +## +# Internal variables. Do not modify this section directly. +# +CHAINS=https://git.zv.io/toolchains; +printf "CHAINS=%s\n" "${CHAINS}"; + +SYSTEM="-adelie-linux-musl"; # we only target musl on Linux +printf "SYSTEM=%s\n" "${SYSTEM}"; + +NATIVE=$(cc -dumpmachine); # host arch, host libc +printf "NATIVE=%s\n" "${NATIVE}"; + +BUILDS="${NATIVE%%-*}${SYSTEM}"; # host arch, musl libc +printf "BUILDS=%s\n" "${BUILDS}"; + +TARGET="${m%:*}${SYSTEM}${m#*:}"; # ultimate Adélie Linux target +printf "TARGET=%s\n" "${TARGET}"; + +MTOOLS=${MCMTOOLS:-"${BASE}/mcmtools"}; # CAREFUL! MAY CAUSE DATA LOSS! +printf "MTOOLS=%s\n" "${MTOOLS}"; + +## +# Default 'PATH' for use inside various rootfs environments. +# +DEF_PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"; + +mkdir -p "${BASE}"; + + +#--------------------------------------------------------------- +# mcmtools (seed) + +## +# Allow the user to supply an existing mcmtools installation. It +# is not ideal but we can allow the user to save some CPU cycles +# at the cost of adding new tools to the existing installation. +# +# If they wish to keep the existing mcmtools installation clean, +# the 'binsub' tool can be used to relocate it. A temporary dir +# is used if this environment variable is omitted. Another case +# for providing a custom value is if '/tmp' is mounted weird. +# +if ! test -d "${MTOOLS}"/sys/emus/bin; then # FIXME: no hard code + cd "${BASE}"; + + test -d bootstrap \ + || git clone ${CHAINS}/bootstrap.git; + cd bootstrap; + git checkout 371f5e2624acd89dd05a9bd8c68f369b16dfdde6; + + ## seed rootfs + # + # This will build absolutely everything that is needed to be + # self-reliant, except for some build deps for QEMU. + # + # We copy 'config.mak' from mcmtools bootstrap to the rootfs + # so that when we build "real" toolchain they are the same. + # + DEST="${MTOOLS}" \ + ARCH=${BUILDS} \ + ./bootstrap \ + ; + test -f "${MTOOLS}"/config.mak || \ + cp "${MTOOLS}"/tmp/musl-cross-make/config.mak \ + "${MTOOLS}"/config.mak \ + ; + # cleaning + ( + cd "${MTOOLS}"; + rm -fr tmp; # save 10 GB + ) + + # is any of this actually needed? + ( + cd "${MTOOLS}"/sys; + mkdir -p dev; + mkdir -p proc; + mkdir -p sys; + rm -fr usr; + ln -s . usr; + ) + + ## emulators + # + # Dependencies are built with the mcmtools host toolchain; a + # reason to not force musl here is in the event that these + # cannot be built statically and the host libc is different. + # Our priority is to obtain a functioning 'PRoot' above all. + # + # QEMU itself is built inside an Alpine Linux rootfs; we do + # this because we still need Python 3 to build it. You can + # manually provide your own static QEMU user binaries and be + # on your way without Alpine, but it is a good 'PRoot' test. + # + test -d "${MTOOLS}/sys/emus/bin" || \ + PATH="${MTOOLS}/host/bin:${MTOOLS}/sys/bin" \ + DEST="${MTOOLS}" \ + ./prootemu \ + ; +fi + + +#--------------------------------------------------------------- +# musl toolchains + +## +# We have a musl-targeting toolchain that runs on the host, but +# it is linked to the host libc and may not run in this chroot. +# +# We need to build the same toolchain, but static. There are two +# possible directions to go: (1) a cross toolchain that runs at +# full speed on the build machine, at the cost of having to tell +# downstream build scripts how to cross compile, or (2) a native +# toolchain for the foreign (target) architecture that runs slow +# in QEMU and still requires the cross toolchain to build it. +# +# Build both. We will eventually need both toolchains, anyway. +# +if ! test -d "${MTOOLS}"/sys/tc/native; then # FIXME: no hard code + cd "${BASE}"; + + test -d musl-cross-make \ + || git clone ${CHAINS}/musl-cross-make.git; + cd musl-cross-make; + git pull; # always use the latest + + ## + # Ensure consistent 'config.mak' for all toolchain builds. + # + cp "${MTOOLS}"/config.mak config.mak; + + ## musl toolchains + # + # Build these toolchains statically using the musl toolchain + # from the seed rootfs so that it is known to work correctly + # (the original musl toolchain itself may itself be linked + # with glibc or be unsafe to use in some contexts). + # + # Note: "native" is for the foreign target CPU architecture. + # + MCMTOOLS="${MTOOLS}" \ + ./scripts/build ${TARGET} \ + ; + for k in cross native; do + rm -fr "${MTOOLS}"/sys/tc/${k}; + mkdir "${MTOOLS}"/sys/tc/${k}; + tar -C "${MTOOLS}"/sys/tc/${k} \ + --strip-components=1 \ + -xzf output/${TARGET}-${k}.tgz \ + ; + done +fi + + +#--------------------------------------------------------------- +# build tools (cross) + +## +# Build 'abuild', its dependencies, and other utilities. +# +PROOT_NO_SECCOMP=1 \ +PATH="/tc/cross/bin:${DEF_PATH}" \ +SHELL=/bin/sh \ +DEST=/usr/local \ +CURL_CA_BUNDLE=/cacert.pem \ +CC=${TARGET}-gcc \ +CXX=${TARGET}-g++ \ +AR=${TARGET}-ar \ +LD=${TARGET}-ld \ +CPP=${TARGET}-cpp \ +RANLIB=${TARGET}-ranlib \ +${MTOOLS}/sys/emus/bin/proot \ + -S "${MTOOLS}"/sys \ + -q "${MTOOLS}"/sys/emus/bin/qemu-${q} \ + -b "${HERE}" \ + "${HERE}"/setup-abuild \ + ; + + +#--------------------------------------------------------------- +# rootfs: build + +## +# Create a tarball of the build rootfs. The image creator could +# use this as input if the target architecture matches. +# +cd "${BASE}"; +if ! test -f rootfs-${BUILDS}.tgz; then + tar -C mcmtools/sys \ + -pczf rootfs-${BUILDS}.tgz \ + . \ + ; +fi + + +#--------------------------------------------------------------- +# mcmtools (host) + +## +# NOTE: The 'PATH' order is *really* important. Foreign "native" +# toolchains must be the first ones found; e.g. '/usr/bin/gcc' +# is a symlink to 'ccache' and isn't a functional compiler. +# +# NOTE: 'LD_LIBRARY_PATH' is needed because of nonstandard path +# of toolchain, which otherwise produces these errors: +# +# Error loading shared library libstdc++.so.6: +# Exec format error (needed by ...) +# Error loading shared library libgcc_s.so.1: +# Exec format error (needed by ...) +# +# NOTE: We *do* need to build 'musl-cross-make' (even though it +# may seem redundant) so that we can safely install the dynamic +# loader. If we do not do this, we're on the hook to symlink it. +# +# Once 'coreutils' is built, 'uname' will return correctly, then +# future software should(tm) build as if it is built natively. +# +# FIXME: is Linux 3.2.0 really appropriate here? +# +cd "${BASE}"; +if ! test -d mcmtools-${TARGET}/sys/tc/musl/bin; then # FIXME: no hard code +PROOT_NO_SECCOMP=1 \ +PATH="/tc/native/bin:${DEF_PATH}" \ +LD_LIBRARY_PATH=/tc/native/lib \ +SHELL=/bin/sh \ +BASE="${BASE}" \ +DEST="${BASE}"/mcmtools-${TARGET} \ +ARCH=${TARGET} \ +CURL_CA_BUNDLE=/cacert.pem \ +${MTOOLS}/sys/emus/bin/proot \ + -S "${MTOOLS}"/sys \ + -q "${MTOOLS}"/sys/emus/bin/qemu-${q} \ + -b "${HERE}" \ + -k "3.2.0" \ + "${BASE}"/bootstrap/bootstrap \ + ; +fi + +# cleaning +( + cd mcmtools-${TARGET}; + rm -fr tmp; # save 10 GB +) + +## +# Install the native static toolchain over the native rootfs for +# a quick way to have the dynamic loader. Overwrite the symlink! +# +if test -h mcmtools-${TARGET}/sys/bin/gcc; then # FIXME: no hard code + tar -C mcmtools-${TARGET}/sys \ + --strip-components=1 \ + -xzf musl-cross-make/output/${TARGET}-native.tgz \ + ; +fi + + +#--------------------------------------------------------------- +# build tools (host) + +## +# Build 'abuild', its dependencies, and other utilities. +# +PROOT_NO_SECCOMP=1 \ +PATH="/tc/musl/bin:${DEF_PATH}" \ +SHELL=/bin/sh \ +DEST=/usr/local \ +CURL_CA_BUNDLE=/cacert.pem \ +CC=gcc \ +CXX=g++ \ +AR=ar \ +LD=ld \ +CPP=cpp \ +${MTOOLS}/sys/emus/bin/proot \ + -S "${BASE}"/mcmtools-${TARGET}/sys \ + -q "${MTOOLS}"/sys/emus/bin/qemu-${q} \ + -b "${HERE}" \ + "${HERE}"/setup-abuild \ + ; + + +#--------------------------------------------------------------- +# rootfs: host + +## +# Create a tarball of the host rootfs. This is a safety measure. +# +cd "${BASE}"; +if ! test -f rootfs-${TARGET}.tgz; then + tar -C mcmtools-${TARGET}/sys \ + -pczf rootfs-${TARGET}.tgz \ + . \ + ; +fi + + +#--------------------------------------------------------------- +# patch rootfs + +## +# Some paths are hard-coded into various binaries. Since we need +# to "relocate" everything to run at '/', we can decompress the +# '.tar' file and perform the appropriate substititons at once. +# +cd "${BASE}"; +if ! test -f rootfs-${TARGET}-patched.tgz; then + gzip -dk rootfs-${TARGET}.tgz; + + ## + # Build 'binsub', a string patch tool. + # + "${MTOOLS}"/musl/bin/gcc -static -o binsub binsub.c -O3; + + ## + # Replace with explicit '/' ('/foo/bar/baz' --> '/' instead + # of the empty string) to avoid gotchas. Assumes '//' is the + # same as '/'. I can't think of a counterexample. + # + ./binsub rootfs-${TARGET}.tar \ + "${BASE}"/mcmtools-${TARGET}/sys \ + / \ + ; + + gzip -c9 rootfs-${TARGET}.tar > rootfs-${TARGET}-patched.tgz; + + rm -f rootfs-${TARGET}.tar; +fi + + +#--------------------------------------------------------------- +# preimage + +## +# The preimage has some modifications to support using the real +# "bootstrap.sh" script. Note that we need to bind-mount and use +# a different working directory because PRoot has a few defaults +# that would otherwise shadow paths we're trying to create. +# +cd "${BASE}"; +if ! test -f rootfs-${TARGET}-preimage.tgz; then # FIXME: no hard code + rm -fr rootfs-${TARGET}-preimage; + mkdir rootfs-${TARGET}-preimage; + tar -C rootfs-${TARGET}-preimage \ + -xzf rootfs-${TARGET}-patched.tgz \ + || true; # we're somehow messing up some bzip binaries? + +PROOT_NO_SECCOMP=1 \ +PATH="${DEF_PATH}" \ +SHELL=/bin/sh \ +CURL_CA_BUNDLE=/cacert.pem \ +${MTOOLS}/sys/emus/bin/proot \ + -S "${MTOOLS}"/sys \ + -b "${BASE}"/rootfs-${TARGET}-preimage \ + -w "${BASE}"/rootfs-${TARGET}-preimage \ + -b "${HERE}" \ + "${HERE}"/setup-preimage \ + ; + + tar -C rootfs-${TARGET}-preimage \ + -pczf rootfs-${TARGET}-preimage.tgz \ + . \ + ; +fi + + +#--------------------------------------------------------------- +# packages + +## +# This is where the original "bootstrap.sh" script runs. The aim +# of this step is to produce '.apk' files that were built using +# non-Adélie-packaged tools, but would install Adélie tools. +# +if ! test -f rootfs-${TARGET}-packages.tgz; then # FIXME: no hard code + rm -fr rootfs-${TARGET}-packages; + mkdir rootfs-${TARGET}-packages; + tar -C rootfs-${TARGET}-packages \ + -xzf rootfs-${TARGET}-preimage.tgz \ + ; + +PROOT_NO_SECCOMP=1 \ +PATH="${DEF_PATH}" \ +SHELL=/bin/sh \ +TARGET_ARCH=${TARGET} \ +CURL_CA_BUNDLE=/cacert.pem \ +${MTOOLS}/sys/emus/bin/proot \ + -R "${BASE}"/rootfs-${TARGET}-packages \ + -i 1000:300 \ + -q "${MTOOLS}"/sys/emus/bin/qemu-${q} \ + -b "${HERE}" \ + "${HERE}"/setup-packages \ + ; +fi + +# TODO: create tarball of this directory + + +#--------------------------------------------------------------- +# image creator + +# TODO + +# apk --root /foo --arch armv7 --initdb add +# apk --root /foo add adelie-core dash-binsh build-tools diff --git a/setup-abuild b/setup-abuild new file mode 100755 index 0000000..686df6d --- /dev/null +++ b/setup-abuild @@ -0,0 +1,362 @@ +#!/bin/sh -e + +HERE="$(dirname $(readlink -f ${0}))"; +DEST=/usr/local; + +git config --global http.sslCAInfo "${CURL_CA_BUNDLE}"; + + +mkdir -p "${DEST}"; +cd "${DEST}"; # this directory will already exist if correct + +## +# musl +# +# This provides the dynamic loader for the foreign (target) arch +# so that we do not have to force building static musl binaries. +# +nmus=musl; +vmus=1.2.3; +test ! -f ._${nmus}-${vmus} && \ +( + test ! -d ${nmus}-${vmus} \ + && curl -s https://musl.libc.org/releases/${nmus}-${vmus}.tar.gz \ + | tar -xzf - \ + ; + cd ${nmus}-${vmus}; + rm -fr x; mkdir x; cd x; + ../configure \ + --prefix=/usr \ + --enable-static \ + --enable-shared \ + ; + make -j$(nproc); + make install; +) +touch ._${nmus}-${vmus}; +rm -fr ${nmus}-${vmus}; + + +## +# OpenSSL +# +nssl=openssl; +vssl=1.1.1s; +test ! -f ._${nssl}-${vssl} && \ +( + test ! -d ${nssl}-${vssl} \ + && curl -s https://www.openssl.org/source/${nssl}-${vssl}.tar.gz \ + | tar -xzf - \ + ; + cd ${nssl}-${vssl}; + rm -fr x; mkdir x; cd x; + ../Configure cc \ + --prefix="${DEST}" \ + --openssldir="${DEST}" \ + no-shared \ + ; + make -j$(nproc); + make install_sw install_ssldirs; +) +touch ._${nssl}-${vssl}; +rm -fr ${nssl}-${vssl}; + + +## +# zlib +# +nzlb=zlib; +vzlb=1.2.13; +test ! -f ._${nzlb}-${vzlb} && \ +( + test ! -d ${nzlb}-${vzlb} \ + && curl -s https://www.zlib.net/${nzlb}-${vzlb}.tar.gz \ + | tar -xzf - \ + ; + cd ${nzlb}-${vzlb}; + rm -fr x; mkdir x; cd x; + ../configure \ + --prefix="${DEST}" \ + --static \ + ; + make -j$(nproc); + make install; +) +touch ._${nzlb}-${vzlb}; +rm -fr ${nzlb}-${vzlb}; + + +## +# abuild +# +nbld=abuild; +vbld=f75ebd1953f6e41416b5037c4049b55b9a21a1ea; +test ! -f ._${nbld}-${vbld} && \ +( + test ! -d ${nbld}-${vbld} \ + && git clone https://git.adelielinux.org/adelie/${nbld}.git ${nbld}-${vbld} \ + ; + cd abuild-${vbld}; + git checkout ${vbld}; + while read k; do curl -s ${k} | patch -p1 || true; done <<EOF +http://ix.io/4ifN +EOF + rm -fr x; mkdir x; cd x; + export SSL_CFLAGS="-I${DEST}/include"; + export SSL_LDFLAGS="-L${DEST}/lib"; + export SSL_LIBS="-lssl -lcrypto"; # not in mcmtools + export ZLIB_LIBS="-lz"; # do not use from mcmtools + export LDFLAGS="-L${DEST}/lib -lssl -lcrypto"; + export CFLAGS="-static -I${DEST}/include"; + sed -i "${DEST}/abuild-${vbld}/abuild-sudo.c" \ + -e "s@__DEST__@${DEST}@" \ + ; # hardcoded + make -j$(nproc) -C .. install \ + prefix="${DEST}" \ + sysconfdir="${DEST}" \ + SCDOC=true \ + ; + sed -i "${DEST}/bin/abuild" \ + -e 's@/bin/ash -e@/usr/bin/env bash@' \ + ; # hardcoded +) +touch ._${nbld}-${vbld}; +rm -fr ${nbld}-${vbld}; + + +## +# util-linux (for 'getopt' used by 'abuild-keygen') +# +nutl=util-linux; +vutl=08431acdf5b3accd0887ab550bfa4efabed751d6; +test ! -f ._${nutl}-${vutl} && \ +( + test ! -d ${nutl}-${vutl} \ + && mkdir ${nutl}-${vutl} \ + && git clone https://github.com/karelzak/${nutl}.git ${nutl}-${vutl} \ + ; + cd ${nutl}-${vutl}; + git checkout ${vutl}; + test -f configure || ./autogen.sh; + rm -fr x; mkdir x; cd x; + ../configure \ + --prefix="${DEST}" \ + --host="$(${CC} -dumpmachine)" \ + --enable-static \ + --disable-shared \ + ; + sed -i Makefile \ + -e 's/chgrp/-chgrp/g' \ + -e 's/chmod/-chmod/g' \ + -e 's/chown/-chown/g' \ + ; # allow non-root installation + make -j$(nproc) install; +) +touch ._${nutl}-${vutl}; +rm -fr ${nutl}-${vutl}; + + +## +# pkgconf (pkg-config replacement) +# +npkg=pkgconf; +vpkg=623b8f7851648a5c476de904a8ffed7b7b679aab; # until autoconf 2.71 +test ! -f ._${npkg}-${vpkg} && +( + test ! -d ${npkg}-${vpkg} \ + && git clone https://github.com/${npkg}/${npkg}.git ${npkg}-${vpkg} \ + ; + cd ${npkg}-${vpkg}; + git checkout ${vpkg}; + ./autogen.sh; + ./configure \ + --prefix="${DEST}" \ + --host="$(${CC} -dumpmachine)" \ + --enable-static \ + --disable-shared \ + --with-system-libdir=/lib:/usr/lib \ + --with-system-includedir=/usr/include \ + ; + make -j$(nproc) install; + ln -s pkgconf "${DEST}"/bin/pkg-config +) +touch ._${npkg}-${vpkg}; +rm -fr ${npkg}-${vpkg}; + + +## +# samurai (ninja replacement) +# +nsam=samurai; +vsam=4cc8f4a3654b72c977e0e62367d1ab6d5b873def; +test ! -f ._${nsam}-${vsam} && +( + test ! -d ${nsam}-${vsam} \ + && git clone https://github.com/michaelforney/${nsam}.git ${nsam}-${vsam} \ + ; + cd ${nsam}-${vsam}; + make -j$(nproc) install \ + PREFIX="" \ + DESTDIR="${DEST}" \ + ; + +) +touch ._${nsam}-${vsam}; +rm -fr ${nsam}-${vsam}; + + +## +# muon (meson replacement) +# +nmun=muon; +vmun=62ce4561b4444e5020dc39aad0381655afeda0d6; +test ! -f ._${nmun}-${vmun} && +( + test ! -d ${nmun}-${vmun} \ + && git clone https://git.sr.ht/~lattis/${nmun} ${nmun}-${vmun} \ + ; + cd ${nmun}-${vmun}; + sed -i bootstrap.sh \ + -e 's/if.*then/if false; then/g' \ + ; + ./bootstrap.sh \ + bootstrap \ + ; + bootstrap/muon setup \ + -Dstatic=true \ + build \ + ; + sed -i build/build.ninja \ + -e "s@\bar\b@${AR}@g" \ + ; + samu -C build; + cp build/muon "${DEST}/bin"; +) +touch ._${nmun}-${vmun}; +rm -fr ${nmun}-${vmun}; + + +## +# apk-tools +# +# Ariadne says use meson, and 'muon' doesn't work so... kludges! +# Also, '-j' will break the build. +# +natl=apk-tools; +vatl=be4ce40797af9056c79be4dc74ff978f1f4957e4; +test ! -f ._${natl}-${vatl} && \ +( + test ! -d ${natl}-${vatl} \ + && git clone https://git.alpinelinux.org/${natl} ${natl}-${vatl} \ + ; + cd ${natl}-${vatl}; + sed -i Make.rules \ + -e '/targets += $(__shlibs) $(shobjs)/d' \ + ; # disable shared libs + sed -i Make.rules \ + -e '/CC.*:=/d' \ + -e '/AR.*:=/d' \ + -e '/LD.*:=/d' \ + ; # inherit from environment + sed -i src/Makefile \ + -e 's/$(install-libapk_so)//g' -e 's/$(libapk_so)//g' \ + -e 's/ version.o/ version.o strlcpy.o/' \ + ; # disable shared libs, hack the hack + sed -i src/context.c \ + -e "s@var/log@${DEST}/${1}/var/log@" \ + ; # hardcoded + ln -sf ../portability/strlcpy.c src/strlcpy.c; + export LUA=no; # documentation requires lua + make clean; + make \ + INSTALLDIR="${DEST}" \ + CFLAGS="-I${DEST}/include -DNEED_STRLCPY -Wno-error" \ + LDFLAGS="-L${DEST}/lib -L${DEST}/${natl}-${vatl}/libfetch" \ + LIBS="-lapk -lfetch -lssl -lcrypto -lz" \ + ; + cp src/apk "${DEST}/bin"; +) +touch ._${natl}-${vatl}; +rm -fr ${natl}-${vatl}; + + +## +# pax-utils +# +npax=pax-utils; +vpax=974b9359c2f89d57e69598572aafcd8f920d79e2; +test ! -f ._${npax}-${vpax} && \ +( + test ! -d ${npax}-${vpax} \ + && git clone https://anongit.gentoo.org/git/proj/${npax}.git ${npax}-${vpax} \ + ; + cd ${npax}-${vpax}; + git checkout ${vpax}; + muon setup build; + samu -C build; + muon -C build install; +) +touch ._${npax}-${vpax}; +rm -fr ${npax}-${vpax}; + + +## +# fakeroot +# +nfrt=fakeroot; +vfrt=8c0260009e85264fd1ea282fbb22063fc694c552; # until autoconf 2.71 +test ! -f ._${nfrt}-${vfrt#*:} && \ +( + test ! -d ${nfrt}-${vfrt} \ + && git clone https://salsa.debian.org/clint/${nfrt}.git ${nfrt}-${vfrt} \ + ; + cd ${nfrt}-${vfrt}; + git checkout ${vfrt}; + while read k; do curl -s ${k} | patch -p1 || true; done <<EOF +https://git.alpinelinux.org/aports/plain/main/fakeroot/do-not-redefine-id_t.patch?id=bb497eeb2155d0332284942105692bc05fec25a9 +https://git.alpinelinux.org/aports/plain/main/fakeroot/fakeroot-no64.patch?id=bb497eeb2155d0332284942105692bc05fec25a9 +https://git.alpinelinux.org/aports/plain/main/fakeroot/fakeroot-stdint.patch?id=bb497eeb2155d0332284942105692bc05fec25a9 +https://git.alpinelinux.org/aports/plain/main/fakeroot/fix-format.patch?id=bb497eeb2155d0332284942105692bc05fec25a9 +https://git.alpinelinux.org/aports/plain/main/fakeroot/fix-shell-in-fakeroot.patch?id=bb497eeb2155d0332284942105692bc05fec25a9 +EOF + ./bootstrap; + f=$(mktemp); # needed due to "error: unknown type name 'cap_user_header_t'" + echo > ${f} "#include <linux/capability.h>"; + cat libfakeroot.c >> ${f}; + mv ${f} libfakeroot.c; + rm -fr x; mkdir x; cd x; + CFLAGS="-D_STAT_VER=0 $CFLAGS" \ + ../configure \ + --prefix="${DEST}" \ + ; + sed -i Makefile \ + -e '/^SUBDIRS =/ s/doc//g' \ + ; + make -j$(nproc) install; +) +touch ._${nfrt}-${vfrt#*:}; +rm -fr ${nfrt}-${vfrt#*:}; + + +## +# attr +# +natt=attr; +vatt=2979615e71fb53b3f2f9085eea516d4e2b3174ea; +test ! -f ._${natt}-${vatt} && +( + test ! -d ${natt}-${vatt} \ + && git clone https://git.savannah.nongnu.org/git/${natt}.git ${natt}-${vatt} \ + ; + cd ${natt}-${vatt}; + git checkout ${vatt}; + ./autogen.sh; + ./configure \ + --prefix="${DEST}" \ + --enable-static \ + --disable-shared \ + ; + make -j$(nproc) install; +) +touch ._${natt}-${vatt}; +rm -fr ${natt}-${vatt}; diff --git a/setup-packages b/setup-packages new file mode 100755 index 0000000..ea449dc --- /dev/null +++ b/setup-packages @@ -0,0 +1,13 @@ +#!/bin/sh -e + +d=$(mktemp -d); +cd "${d}"; +pwd; + +git clone https://git.adelielinux.org/adelie/packages.git; +cd packages; +git checkout 198e0130d845f7466c09a61f3829905d95313c06; + +# FIXME: this script needs a lot of work +#./scripts/bootstrap.sh ${TARGET_ARCH}; +bash diff --git a/setup-preimage b/setup-preimage new file mode 100755 index 0000000..7892a0d --- /dev/null +++ b/setup-preimage @@ -0,0 +1,35 @@ +#!/bin/sh -e + +set -x; + +# these will require proper mounting when actually used +mkdir -p dev proc sys tmp; + +# overwrite leftovers from mcmtools bootstrap +cat > root/.bashrc <<EOF +PATH="${PATH}"; +EOF + +# handle git backend issue (hardcoded paths) +# git config --global http.sslCAInfo /cacert.pem +cat > root/.gitconfig <<EOF +[http] + sslCAInfo = "${CURL_CA_BUNDLE}" +EOF + +# handle CA certificate issue (this or 'CURL_CA_BUNDLE' env var) +cat > root/.curlrc <<EOF +--cacert "${CURL_CA_BUNDLE}" +EOF + +# hack to create abuild user and group +# FIXME: have to use a patched 'abuild' anyway, debug me! +cat > etc/group <<EOF +abuild:x:300:builder +EOF +cat > etc/passwd <<EOF +builder:x:1000:300::/usr/src:/bin/sh +EOF + +# maybe not necessary? +chown -R 1000:300 root etc; |