commit 15391ae704f949821f6f8e7f7088a106e12b18d0
parent 16e81e5d753312600d9beaac52a9155841662cf0
Author: Ryan Sepassi <rsepassi@gmail.com>
Date: Wed, 6 May 2026 14:32:09 -0700
B (scripts): unify bootN.sh; per-stage + cumulative timings
Sweep the scripts/boot{,0..6}.sh + lib-{arch,pipeline,runscm}.sh
surface for clarity and uniformity. No behavior change to the build
chain itself.
lib-arch.sh:
- bootlib_init exports BOOT_STAGE; driver_init now sets OUT/STAGE
automatically (was duplicated per script).
- Add require_src and require_file <path> [<hint>] helpers,
replacing seven copies of `[ -d "$SRC" ] || ...` and several
ad-hoc file checks.
- bootlib_init records BOOT_T0 and installs an EXIT trap; on
success every stage prints `[BOOT_TAG] done in Xs (cum Ys)`
where cum sums all per-stage .timing files for the current
arch/driver. Works for both `make build/.../Image` and direct
`boot.sh`.
lib-pipeline.sh, lib-runscm.sh:
- pipeline_export / runscm_export accept varargs.
- *_input_from_src and runscm_input_tree_from_src drop the
leading `_kind` arg (always `src`); the one bin case in boot0
inlines pipeline_input.
- Add runscm_gen <gen-script> <args...>: emits run.scm to
$STAGE/run.scm, logs size, registers it. Replaces a 4-line
pattern in boot4/5/6.
- Document why `stage` needs explicit -- input/output lists.
boot.sh:
- set -e -> set -eu; rename helper stage() -> time_stage(); then
drop time_stage entirely since per-child timing now comes from
each child's lib trap.
boot0..boot6:
- Standardized header docstrings; trimmed boot4's tcc1!=tcc2 essay.
- boot1: build_p1 helper; boot2: build_p1pp helper. The duplicated
4-stage chains are now one call per output.
Verified by `sh -n`, shellcheck, and end-to-end smoke runs of
boot0/1/2 on aarch64/podman with timing output.
Diffstat:
11 files changed, 329 insertions(+), 335 deletions(-)
diff --git a/scripts/boot.sh b/scripts/boot.sh
@@ -14,7 +14,7 @@
## DRIVER=seed ./scripts/boot.sh <arch> # re-run on tcc-built kernel
## Subsequent DRIVER=seed runs reuse the Image directly — no stashing.
-set -e
+set -eu
case "${1:-}" in
-h|--help)
@@ -63,35 +63,28 @@ fi
# (the seed driver consumes build/$ARCH/podman/boot6/$KERNEL_NAME).
rm -rf build/$ARCH/$DRIVER
-T0=$(date +%s)
-trap 'echo "[$BOOT_TAG] elapsed at exit: $(($(date +%s) - T0))s"' EXIT
-
-stage() {
- name=$1; shift
- s=$(date +%s)
- "$@"
- e=$(date +%s)
- echo "[$BOOT_TAG] $name: $((e - s))s (cum $((e - T0))s)"
-}
+# Per-stage timing comes from each child's own bootlib EXIT trap
+# (`[bootN/$DRIVER/$ARCH] done in Xs (cum Ys)`); this orchestrator only
+# adds its own total at the end (also via the lib trap).
# A0a: build the canonical generated source tree at build/$ARCH/src/.
# Boot stages read source from there exclusively (no flatten/unpack/
# patch inside boot{N}.sh).
-stage prep-src ./scripts/prep-src.sh $ARCH
+./scripts/prep-src.sh $ARCH
-stage boot0 ./scripts/boot0.sh $ARCH
-stage boot1 ./scripts/boot1.sh $ARCH
-stage boot2 ./scripts/boot2.sh $ARCH
-stage boot3 ./scripts/boot3.sh $ARCH
-stage boot4 ./scripts/boot4.sh $ARCH
+./scripts/boot0.sh $ARCH
+./scripts/boot1.sh $ARCH
+./scripts/boot2.sh $ARCH
+./scripts/boot3.sh $ARCH
+./scripts/boot4.sh $ARCH
# A0b: apply the per-arch musl skip filter (needs tcc3 from boot4 if
# the calibration list is missing; the committed list is the common
# case and runs without compiler).
-stage prep-musl ./scripts/prep-musl.sh $ARCH
+./scripts/prep-musl.sh $ARCH
-stage boot5 ./scripts/boot5.sh $ARCH
+./scripts/boot5.sh $ARCH
# boot6 builds the seed-kernel ELF/Image with boot4's tcc3 (no `ld -T`,
# no objcopy).
-stage boot6 ./scripts/boot6.sh $ARCH
+./scripts/boot6.sh $ARCH
diff --git a/scripts/boot0.sh b/scripts/boot0.sh
@@ -1,15 +1,14 @@
#!/bin/sh
-## boot0.sh — standalone seed bootstrap.
+## boot0.sh — seed bootstrap: hex0-seed → hex0 → hex1 → hex2 → catm → M0.
##
## Stage 0 of the README's chain. From the ~400-byte vendored hex0-seed,
-## brings up: hex0 -> hex1 -> hex2 -> catm -> M0. Three of those (hex2,
-## catm, M0) are the binaries every later stage depends on.
+## brings up the three binaries every later stage depends on (hex2, catm,
+## M0).
##
## ─── Inputs (sources, from canonical tree) ───────────────────────────
## build/$ARCH/src/bin/hex0-seed
## build/$ARCH/src/src/vendor-seed/{hex0.hex0, hex1.hex0, hex2.hex1,
## catm.hex2, M0.hex2, ELF.hex2}
-## (populated by scripts/prep-src.sh from vendor/seed/$ARCH/.)
##
## ─── Outputs ──────────────────────────────────────────────────────────
## build/$ARCH/$DRIVER/boot0/{hex2, catm, M0}
@@ -22,20 +21,15 @@ set -eu
. scripts/lib-arch.sh
bootlib_init boot0 "${1:-}"
driver_init scratch
-
-SRC=build/$ARCH/src
-OUT=build/$ARCH/$DRIVER/boot0
-STAGE=build/$ARCH/$DRIVER/.boot0-stage
-
-[ -d "$SRC" ] || { echo "[$BOOT_TAG] missing $SRC — run scripts/prep-src.sh $ARCH" >&2; exit 1; }
+require_src
. scripts/lib-pipeline.sh
pipeline_init "$STAGE" "$OUT" "$DRIVER"
# ─── inputs (from canonical src tree) ─────────────────────────────────
-pipeline_input_from_src bin hex0-seed
+pipeline_input hex0-seed "build/$ARCH/src/bin/hex0-seed"
for f in hex0.hex0 hex1.hex0 hex2.hex1 catm.hex2 M0.hex2 ELF.hex2; do
- pipeline_input_from_src src "vendor-seed/$f"
+ pipeline_input_from_src "vendor-seed/$f"
done
# ─── pipeline ─────────────────────────────────────────────────────────
@@ -48,10 +42,7 @@ stage hex2 catm.hex2 catm -- catm.hex2
stage catm M0.combined.hex2 ELF.hex2 M0.hex2 -- ELF.hex2 M0.hex2 -- M0.combined.hex2
stage hex2 M0.combined.hex2 M0 -- M0.combined.hex2 -- M0
-pipeline_export hex2
-pipeline_export catm
-pipeline_export M0
-
+pipeline_export hex2 catm M0
pipeline_run
echo "[$BOOT_TAG] OK -> $OUT/{hex2, catm, M0}"
diff --git a/scripts/boot1.sh b/scripts/boot1.sh
@@ -1,9 +1,9 @@
#!/bin/sh
-## boot1.sh — standalone build of M1pp + hex2pp.
+## boot1.sh — build the self-hosted M1pp + hex2pp pair.
##
-## Stage 1 of the README's chain: produces the self-hosted M1pp +
-## hex2pp pair, built from their .P1 sources via the seed M0 + hex2
-## chain. catm is rebuilt from catm.P1pp in boot2.
+## Stage 1 of the README's chain: produces M1pp and hex2pp from their
+## .P1 sources via the seed M0 + hex2 chain. catm is rebuilt later in
+## boot2 from catm.P1pp.
##
## ─── Inputs (sources, from canonical tree) ───────────────────────────
## build/$ARCH/src/src/M1pp/M1pp.P1
@@ -25,14 +25,10 @@ set -eu
. scripts/lib-arch.sh
bootlib_init boot1 "${1:-}"
driver_init scratch
+require_src
BOOT0=build/$ARCH/$DRIVER/boot0
-SRC=build/$ARCH/src
-OUT=build/$ARCH/$DRIVER/boot1
-STAGE=build/$ARCH/$DRIVER/.boot1-stage
-
require_prev "$BOOT0" hex2 M0 catm
-[ -d "$SRC" ] || { echo "[$BOOT_TAG] missing $SRC — run scripts/prep-src.sh $ARCH" >&2; exit 1; }
. scripts/lib-pipeline.sh
pipeline_init "$STAGE" "$OUT" "$DRIVER"
@@ -41,35 +37,29 @@ pipeline_init "$STAGE" "$OUT" "$DRIVER"
pipeline_input hex2 "$BOOT0/hex2"
pipeline_input M0 "$BOOT0/M0"
pipeline_input catm "$BOOT0/catm"
-pipeline_input_from_src src "P1/P1-$ARCH.M1" P1.M1
-pipeline_input_from_src src vendor-seed/ELF.hex2
-pipeline_input_from_src src M1pp/M1pp.P1
-pipeline_input_from_src src hex2pp/hex2pp.P1
+pipeline_input_from_src "P1/P1-$ARCH.M1" P1.M1
+pipeline_input_from_src vendor-seed/ELF.hex2
+pipeline_input_from_src M1pp/M1pp.P1
+pipeline_input_from_src hex2pp/hex2pp.P1
# ─── pipeline ─────────────────────────────────────────────────────────
-echo "[$BOOT_TAG] M1pp.P1 + hex2pp.P1 -> M1pp + hex2pp"
-
-# .P1 -> ELF via M0 + hex2:
-# catm P1.M1 + src -> combined.M1
+# .P1 -> ELF, applied to each of M1pp.P1 and hex2pp.P1:
+# catm P1.M1 + <src> -> combined.M1
# M0 combined.M1 -> prog.hex2
# catm ELF.hex2 + prog.hex2 -> linked.hex2
# hex2 linked.hex2 -> ELF binary
+build_p1() { # $1 = source .P1, $2 = output binary name
+ stage catm combined.M1 P1.M1 "$1" -- P1.M1 "$1" -- combined.M1
+ stage M0 combined.M1 prog.hex2 -- combined.M1 -- prog.hex2
+ stage catm linked.hex2 ELF.hex2 prog.hex2 -- ELF.hex2 prog.hex2 -- linked.hex2
+ stage hex2 linked.hex2 "$2" -- linked.hex2 -- "$2"
+}
-# M1pp.P1 -> M1pp
-stage catm combined.M1 P1.M1 M1pp.P1 -- P1.M1 M1pp.P1 -- combined.M1
-stage M0 combined.M1 prog.hex2 -- combined.M1 -- prog.hex2
-stage catm linked.hex2 ELF.hex2 prog.hex2 -- ELF.hex2 prog.hex2 -- linked.hex2
-stage hex2 linked.hex2 M1pp -- linked.hex2 -- M1pp
-
-# hex2pp.P1 -> hex2pp
-stage catm combined.M1 P1.M1 hex2pp.P1 -- P1.M1 hex2pp.P1 -- combined.M1
-stage M0 combined.M1 prog.hex2 -- combined.M1 -- prog.hex2
-stage catm linked.hex2 ELF.hex2 prog.hex2 -- ELF.hex2 prog.hex2 -- linked.hex2
-stage hex2 linked.hex2 hex2pp -- linked.hex2 -- hex2pp
-
-pipeline_export M1pp
-pipeline_export hex2pp
+echo "[$BOOT_TAG] M1pp.P1 + hex2pp.P1 -> M1pp + hex2pp"
+build_p1 M1pp.P1 M1pp
+build_p1 hex2pp.P1 hex2pp
+pipeline_export M1pp hex2pp
pipeline_run
echo "[$BOOT_TAG] OK -> $OUT/{M1pp, hex2pp}"
diff --git a/scripts/boot2.sh b/scripts/boot2.sh
@@ -1,10 +1,10 @@
#!/bin/sh
## boot2.sh — rebuild catm via M1pp+hex2pp, then build scheme1.
##
-## Stage 2 of the README's chain: first rebuilds catm from catm.P1pp
-## via the freshly-built M1pp + hex2pp pipeline (replacing the seed
-## boot0 catm so later stages run with zero boot0 dependencies); then
-## builds the scheme1 interpreter from scheme1.P1pp using the new catm.
+## Stage 2 of the README's chain. First rebuilds catm from catm.P1pp via
+## the freshly-built M1pp+hex2pp pipeline (replacing the seed boot0 catm
+## so later stages have zero boot0 dependencies); then builds the
+## scheme1 interpreter from scheme1.P1pp using the new catm.
##
## ─── Inputs (sources, from canonical tree) ───────────────────────────
## build/$ARCH/src/src/catm/catm.P1pp
@@ -27,16 +27,12 @@ set -eu
. scripts/lib-arch.sh
bootlib_init boot2 "${1:-}"
driver_init scratch
+require_src
BOOT0=build/$ARCH/$DRIVER/boot0
BOOT1=build/$ARCH/$DRIVER/boot1
-SRC=build/$ARCH/src
-OUT=build/$ARCH/$DRIVER/boot2
-STAGE=build/$ARCH/$DRIVER/.boot2-stage
-
require_prev "$BOOT0" catm
require_prev "$BOOT1" M1pp hex2pp
-[ -d "$SRC" ] || { echo "[$BOOT_TAG] missing $SRC — run scripts/prep-src.sh $ARCH" >&2; exit 1; }
. scripts/lib-pipeline.sh
pipeline_init "$STAGE" "$OUT" "$DRIVER"
@@ -45,39 +41,33 @@ pipeline_init "$STAGE" "$OUT" "$DRIVER"
pipeline_input catm0 "$BOOT0/catm" # bootstrap; replaced by output 'catm'
pipeline_input M1pp "$BOOT1/M1pp"
pipeline_input hex2pp "$BOOT1/hex2pp"
-pipeline_input_from_src src "P1/P1-$ARCH.M1pp" backend.M1pp
-pipeline_input_from_src src P1/P1.M1pp frontend.M1pp
-pipeline_input_from_src src P1/P1pp.P1pp libp1pp.P1pp
-pipeline_input_from_src src vendor-seed/ELF.hex2
-pipeline_input_from_src src catm/catm.P1pp
-pipeline_input_from_src src scheme1/scheme1.P1pp
+pipeline_input_from_src "P1/P1-$ARCH.M1pp" backend.M1pp
+pipeline_input_from_src P1/P1.M1pp frontend.M1pp
+pipeline_input_from_src P1/P1pp.P1pp libp1pp.P1pp
+pipeline_input_from_src vendor-seed/ELF.hex2
+pipeline_input_from_src catm/catm.P1pp
+pipeline_input_from_src scheme1/scheme1.P1pp
# ─── pipeline ─────────────────────────────────────────────────────────
-echo "[$BOOT_TAG] catm.P1pp -> catm; scheme1.P1pp -> scheme1"
-
-# .P1pp -> ELF:
-# catm backend + frontend + libp1pp + src -> combined.M1pp
+# .P1pp -> ELF, applied to each of catm.P1pp and scheme1.P1pp:
+# catm backend + frontend + libp1pp + <src> -> combined.M1pp
# M1pp combined.M1pp -> expanded.hex2pp
# catm ELF.hex2 + expanded.hex2pp -> linked.hex2pp
# hex2pp -B 0x600000 linked.hex2pp -> ELF binary
+build_p1pp() { # $1 = catm-bin name (catm0 or catm), $2 = src .P1pp, $3 = out
+ _catm=$1; _src=$2; _out=$3
+ stage "$_catm" combined.M1pp backend.M1pp frontend.M1pp libp1pp.P1pp "$_src" \
+ -- backend.M1pp frontend.M1pp libp1pp.P1pp "$_src" -- combined.M1pp
+ stage M1pp combined.M1pp expanded.hex2pp -- combined.M1pp -- expanded.hex2pp
+ stage "$_catm" linked.hex2pp ELF.hex2 expanded.hex2pp -- ELF.hex2 expanded.hex2pp -- linked.hex2pp
+ stage hex2pp -B 0x600000 linked.hex2pp "$_out" -- linked.hex2pp -- "$_out"
+}
-# catm.P1pp -> catm (bootstrap with boot0 catm)
-stage catm0 combined.M1pp backend.M1pp frontend.M1pp libp1pp.P1pp catm.P1pp \
- -- backend.M1pp frontend.M1pp libp1pp.P1pp catm.P1pp -- combined.M1pp
-stage M1pp combined.M1pp expanded.hex2pp -- combined.M1pp -- expanded.hex2pp
-stage catm0 linked.hex2pp ELF.hex2 expanded.hex2pp -- ELF.hex2 expanded.hex2pp -- linked.hex2pp
-stage hex2pp -B 0x600000 linked.hex2pp catm -- linked.hex2pp -- catm
-
-# scheme1.P1pp -> scheme1 (uses just-built catm)
-stage catm combined.M1pp backend.M1pp frontend.M1pp libp1pp.P1pp scheme1.P1pp \
- -- backend.M1pp frontend.M1pp libp1pp.P1pp scheme1.P1pp -- combined.M1pp
-stage M1pp combined.M1pp expanded.hex2pp -- combined.M1pp -- expanded.hex2pp
-stage catm linked.hex2pp ELF.hex2 expanded.hex2pp -- ELF.hex2 expanded.hex2pp -- linked.hex2pp
-stage hex2pp -B 0x600000 linked.hex2pp scheme1 -- linked.hex2pp -- scheme1
-
-pipeline_export catm
-pipeline_export scheme1
+echo "[$BOOT_TAG] catm.P1pp -> catm; scheme1.P1pp -> scheme1"
+build_p1pp catm0 catm.P1pp catm # bootstrap with boot0 catm
+build_p1pp catm scheme1.P1pp scheme1 # uses just-built catm
+pipeline_export catm scheme1
pipeline_run
echo "[$BOOT_TAG] OK -> $OUT/{catm, scheme1}"
diff --git a/scripts/boot3.sh b/scripts/boot3.sh
@@ -1,11 +1,10 @@
#!/bin/sh
-## boot3.sh — bootstrap tcc0 from cc.scm (Stage A of the four-stage tcc
-## chain; the tcc0 → tcc1 → tcc2 → tcc3 rebuild lives in boot4.sh).
+## boot3.sh — bootstrap tcc0 from cc.scm.
##
-## README's `(define tcc (tcc1 tcc.c))`: cc.scm compiles tcc.flat.c into
-## tcc0. boot3 stops there. boot4 picks up tcc0 and self-hosts the rest
-## of the chain (tcc0 → tcc1 → tcc2 → tcc3, with tcc2 == tcc3 as the
-## fixed-point check).
+## Stage A of the four-stage tcc chain: cc.scm compiles tcc.flat.c into
+## tcc0. boot4 picks up tcc0 and self-hosts the rest of the chain
+## (tcc0 → tcc1 → tcc2 → tcc3, with tcc2 == tcc3 as the fixed-point
+## check).
##
## tcc0 = tcc-source compiled by cc.scm ← produced here
## tcc1 = tcc-source compiled by tcc0 ← boot4
@@ -13,36 +12,30 @@
## tcc3 = tcc-source compiled by tcc2 ← boot4
##
## ─── Inputs (sources, from canonical tree) ───────────────────────────
-## build/$ARCH/src/src/scheme1/{prelude.scm} scheme bundle
-## build/$ARCH/src/src/cc/{cc.scm,main.scm} scheme bundle
-## build/$ARCH/src/src/P1/{P1-$ARCH.M1pp,P1.M1pp,P1pp.P1pp} M1pp pipeline
-## build/$ARCH/src/src/P1/{entry-libc.P1pp,elf-end.P1pp} link framing
-## build/$ARCH/src/src/vendor-seed/ELF.hex2 ELF header
-## build/$ARCH/src/src/tcc/tcc.flat.c flattened tcc TU
-## build/$ARCH/src/src/libc/libc.flat.c flattened mes-libc TU
-## (populated up-front by scripts/prep-src.sh; this stage does
-## no flatten/unpack/patch.)
+## build/$ARCH/src/src/scheme1/prelude.scm scheme bundle
+## build/$ARCH/src/src/cc/{cc.scm, main.scm} scheme bundle
+## build/$ARCH/src/src/P1/{P1-$ARCH.M1pp, P1.M1pp, P1pp.P1pp} M1pp pipeline
+## build/$ARCH/src/src/P1/{entry-libc.P1pp, elf-end.P1pp} link framing
+## build/$ARCH/src/src/vendor-seed/ELF.hex2 ELF header
+## build/$ARCH/src/src/tcc/tcc.flat.c flattened tcc TU
+## build/$ARCH/src/src/libc/libc.flat.c flattened mes-libc TU
##
## ─── Inputs (binaries from prior stages) ──────────────────────────────
-## build/$ARCH/$DRIVER/boot1/{M1pp, hex2pp} — built by scripts/boot1.sh
-## build/$ARCH/$DRIVER/boot2/{catm, scheme1} — built by scripts/boot2.sh
+## build/$ARCH/$DRIVER/boot1/{M1pp, hex2pp}
+## build/$ARCH/$DRIVER/boot2/{catm, scheme1}
##
## ─── Tools ────────────────────────────────────────────────────────────
-## In container: scratch + busybox (no libc, no /etc, no resolver).
-## scheme1 evaluates scripts/boot3-run.scm against the
-## flat staging root (cwd=/work for podman, cwd=/ for
-## the seed kernel). Same run.scm drives both.
-## On host: none — Stage A is pure scheme1 + M1pp + hex2pp; no
-## asm step is required.
+## scheme1 evaluates scripts/boot3-run.scm against the flat staging
+## root. Same run.scm drives both DRIVER=podman (cwd=/work) and
+## DRIVER=seed (cwd=/). Stage A is pure scheme1 + M1pp + hex2pp; no
+## asm step.
##
## ─── Outputs ──────────────────────────────────────────────────────────
-## build/$ARCH/$DRIVER/boot3/tcc0 — cc.scm-built bootstrap tcc,
-## consumed by scripts/boot4.sh
+## build/$ARCH/$DRIVER/boot3/tcc0 — cc.scm-built bootstrap tcc
## build/$ARCH/$DRIVER/boot3/libc.P1pp — cc.scm-built mes-libc (lib mode);
## consumed by the cc-libc test suite
## build/$ARCH/$DRIVER/boot3/tcc.flat.P1pp — cc.scm-built tcc TU (lib mode);
-## retained as a debug/inspection
-## artifact
+## debug/inspection artifact
##
## Usage: scripts/boot3.sh <arch>
## <arch> ∈ {aarch64, amd64, riscv64} for either DRIVER (default podman).
@@ -52,17 +45,14 @@ set -eu
. scripts/lib-arch.sh
bootlib_init boot3 "${1:-}"
driver_init empty
+require_src
BOOT1=build/$ARCH/$DRIVER/boot1
BOOT2=build/$ARCH/$DRIVER/boot2
SRC=build/$ARCH/src
-OUT=build/$ARCH/$DRIVER/boot3
-STAGE=build/$ARCH/$DRIVER/.boot3-stage
-# ── prerequisites ─────────────────────────────────────────────────────
require_prev "$BOOT1" M1pp hex2pp
require_prev "$BOOT2" catm scheme1
-[ -d "$SRC" ] || { echo "[$BOOT_TAG] missing $SRC — run scripts/prep-src.sh $ARCH" >&2; exit 1; }
# ── stage inputs and run scheme1 + boot3-run.scm under $DRIVER ────────
. scripts/lib-runscm.sh
@@ -77,23 +67,21 @@ runscm_input hex2pp "$BOOT1/hex2pp"
# scheme1 binary itself is staged by runscm_run (so a `(run "scheme1" …)`
# inside boot3-run.scm finds it at cwd-relative ./scheme1).
-runscm_input_from_src src scheme1/prelude.scm
-runscm_input_from_src src cc/cc.scm
-runscm_input_from_src src cc/main.scm
+runscm_input_from_src scheme1/prelude.scm
+runscm_input_from_src cc/cc.scm
+runscm_input_from_src cc/main.scm
-runscm_input_from_src src "P1/P1-$ARCH.M1pp" backend.M1pp
-runscm_input_from_src src P1/P1.M1pp frontend.M1pp
-runscm_input_from_src src P1/P1pp.P1pp libp1pp.P1pp
-runscm_input_from_src src P1/entry-libc.P1pp
-runscm_input_from_src src P1/elf-end.P1pp
-runscm_input_from_src src vendor-seed/ELF.hex2
+runscm_input_from_src "P1/P1-$ARCH.M1pp" backend.M1pp
+runscm_input_from_src P1/P1.M1pp frontend.M1pp
+runscm_input_from_src P1/P1pp.P1pp libp1pp.P1pp
+runscm_input_from_src P1/entry-libc.P1pp
+runscm_input_from_src P1/elf-end.P1pp
+runscm_input_from_src vendor-seed/ELF.hex2
-runscm_input_from_src src tcc/tcc.flat.c
-runscm_input_from_src src libc/libc.flat.c
+runscm_input_from_src tcc/tcc.flat.c
+runscm_input_from_src libc/libc.flat.c
-runscm_export tcc0
-runscm_export libc.P1pp
-runscm_export tcc.flat.P1pp
+runscm_export tcc0 libc.P1pp tcc.flat.P1pp
runscm_run "${BOOT3_TIMEOUT:-1800}"
echo "[$BOOT_TAG] sizes: tcc0=$(wc -c <"$OUT/tcc0") libc.P1pp=$(wc -c <"$OUT/libc.P1pp")"
diff --git a/scripts/boot4.sh b/scripts/boot4.sh
@@ -5,14 +5,8 @@
## the four-stage chain: tcc0 → tcc1 → tcc2 → tcc3. The bootstrap
## fixed-point check is `tcc2 == tcc3`: once tcc is compiling itself
## with no help from cc.scm, the chain reaches a byte-identical fixed
-## point. tcc0 ≠ tcc1 in *behavior* (not just in code size) because
-## cc.scm's emitted machine code introduces subtle codegen-decision
-## differences — e.g. on riscv64 cc.scm misses several immediate-folding
-## peepholes that tcc applies, so tcc0(tcc.flat.c) emits ~200 more bytes
-## of `.text` than tcc1(tcc.flat.c) does. tcc1 is faithful tcc behavior
-## (its source is tcc.flat.c, run through the cc.scm-built tcc0
-## translator semantically intact); tcc2 is the first binary whose
-## machine code was emitted by faithful tcc.
+## point. (See docs/PLAN.md for the cc.scm vs tcc codegen-divergence
+## reasoning behind needing four stages rather than two.)
##
## tcc0 = tcc-source compiled by cc.scm ← boot3
## tcc1 = tcc-source compiled by tcc0 ← produced here
@@ -20,53 +14,45 @@
## tcc3 = tcc-source compiled by tcc2 ← produced here
##
## ─── Inputs (sources, from canonical tree) ───────────────────────────
-## build/$ARCH/src/src/tcc-libc/$ARCH/{start.S,sys_stubs.S}
-## — _start, sys_* syscall wrappers
-## build/$ARCH/src/src/tcc-cc/mem.c memcpy/memmove/memset/memcmp
-## build/$ARCH/src/src/tcc/tcc-0.9.26-1147-gee75a10c/lib/libtcc1.c
-## (amd64: generic compiler helper runtime)
-## build/$ARCH/src/src/tcc/tcc-0.9.26-1147-gee75a10c/lib/lib-arm64.c
-## (aarch64 + riscv64: TFmode soft-float)
-## build/$ARCH/src/src/tcc/tcc-0.9.26-1147-gee75a10c/lib/va_list.c
-## (amd64: __va_start / __va_arg)
-## build/$ARCH/src/src/tcc/tcc-0.9.26-1147-gee75a10c/lib/alloca86_64*.S
-## (amd64: alloca helpers)
-## build/$ARCH/src/src/tcc/tcc.flat.c flattened tcc TU
-## build/$ARCH/src/src/libc/libc.flat.c flattened mes-libc TU
-## build/$ARCH/src/src/test-fixtures/boot-hello.c smoke binary
-## (populated up-front by scripts/prep-src.sh; this stage does
-## no flatten/unpack/patch.)
+## build/$ARCH/src/src/tcc-libc/$ARCH/{start.S, sys_stubs.S}
+## build/$ARCH/src/src/tcc-cc/mem.c
+## build/$ARCH/src/src/tcc/tcc-0.9.26-1147-gee75a10c/lib/<arch-specific>
+## build/$ARCH/src/src/tcc/tcc.flat.c
+## build/$ARCH/src/src/libc/libc.flat.c
+## build/$ARCH/src/src/test-fixtures/boot-hello.c
##
## ─── Inputs (binaries from prior stages) ──────────────────────────────
-## build/$ARCH/$DRIVER/boot3/tcc0 — built by scripts/boot3.sh
+## build/$ARCH/$DRIVER/boot3/tcc0
+## build/$ARCH/$DRIVER/boot2/{catm, scheme1}
##
## ─── Tools ────────────────────────────────────────────────────────────
-## In container: scratch + busybox (no libc, no /etc, no resolver).
-## scheme1 evaluates a host-generated run.scm (emitted
-## by scripts/boot4-gen-runscm.sh) against the flat
-## staging root (cwd=/work for podman, cwd=/ for the
-## seed kernel). Same run.scm drives both.
-## On host: none — every arch has CONFIG_TCC_ASM and assembles
-## .S inputs (start.S, sys_stubs.S) directly inside the
-## container in stages B/D/E. The aarch64 assembler is
-## the phase-1 arm64-asm.c that flatten patches into
-## tcc-0.9.26 (see docs/TCC-ARM64-ASM.md).
+## scheme1 evaluates a host-generated run.scm (from boot4-gen-runscm.sh)
+## against the flat staging root. Every arch has CONFIG_TCC_ASM and
+## assembles .S inputs (start.S, sys_stubs.S) directly inside the
+## container; no host asm step. The aarch64 assembler is the phase-1
+## arm64-asm.c that flatten patches into tcc-0.9.26 (see
+## docs/TCC-ARM64-ASM.md).
##
## ─── Outputs ──────────────────────────────────────────────────────────
-## build/$ARCH/$DRIVER/boot4/tcc1 — tcc0-built tcc (stage-2 in tests)
-## build/$ARCH/$DRIVER/boot4/tcc2 — tcc1-built tcc (stage-3 in tests)
-## build/$ARCH/$DRIVER/boot4/tcc3 — final fixed-point self-host tcc
+## build/$ARCH/$DRIVER/boot4/{tcc1, tcc2, tcc3}
+## tcc2 and tcc3 are byte-identical (asserted
+## below) — that equality is the fixed-point.
## build/$ARCH/$DRIVER/boot4/crt1.o
-## — tcc2-built startup object, kept outside
-## libc.a because it must lead link lines
+## tcc2-built startup object, kept outside
+## libc.a because it must lead link lines.
## build/$ARCH/$DRIVER/boot4/libc.a
-## — tcc2-built archive of sys_stubs.o + mem.o
-## + libc.o
+## tcc2-built archive of sys_stubs.o + mem.o
+## + libc.o
## build/$ARCH/$DRIVER/boot4/libtcc1.a
-## — tcc2-built tcc compiler helper archive
+## tcc2-built tcc compiler helper archive
## build/$ARCH/$DRIVER/boot4/hello — mes-libc-linked smoke binary
-## tcc2 and tcc3 are byte-identical (asserted at the end of this
-## script) — that equality is the fixed-point check.
+##
+## ─── Env knobs ────────────────────────────────────────────────────────
+## TCC_BOOTSTRAP_RELAX_FIXEDPOINT=1
+## After a codegen-altering tcc patch, the two-stage rule needs a
+## third bounce to converge. Set this to accept tcc3 even when
+## tcc2 != tcc3; the next boot4 run, started from this run's
+## tcc3, will reach tcc2 == tcc3 with no extra knob.
##
## Usage: scripts/boot4.sh <arch>
## <arch> ∈ {aarch64, amd64, riscv64} for either DRIVER (default podman).
@@ -76,6 +62,7 @@ set -eu
. scripts/lib-arch.sh
bootlib_init boot4 "${1:-}"
driver_init empty
+require_src
case "$ARCH" in
aarch64) LIBTCC1_C_SRCS="lib-arm64.c"; LIBTCC1_ASM_SRCS="" ;;
@@ -86,8 +73,6 @@ esac
BOOT2=build/$ARCH/$DRIVER/boot2
BOOT3=build/$ARCH/$DRIVER/boot3
SRC=build/$ARCH/src
-OUT=build/$ARCH/$DRIVER/boot4
-STAGE=build/$ARCH/$DRIVER/.boot4-stage
TCC_PKG=tcc-0.9.26-1147-gee75a10c
TCC_LIB_REL=tcc/$TCC_PKG/lib
@@ -95,54 +80,36 @@ TCC_LIB_REL=tcc/$TCC_PKG/lib
# ── prerequisites ─────────────────────────────────────────────────────
require_prev "$BOOT3" tcc0
require_prev "$BOOT2" catm scheme1
-[ -d "$SRC" ] || { echo "[$BOOT_TAG] missing $SRC — run scripts/prep-src.sh $ARCH" >&2; exit 1; }
for f in $LIBTCC1_C_SRCS $LIBTCC1_ASM_SRCS; do
- [ -e "$SRC/src/$TCC_LIB_REL/$f" ] || { echo "[$BOOT_TAG] missing $SRC/src/$TCC_LIB_REL/$f" >&2; exit 1; }
+ require_file "$SRC/src/$TCC_LIB_REL/$f"
done
# ── stage inputs and run scheme1 + boot4 run.scm under $DRIVER ────────
. scripts/lib-runscm.sh
runscm_init "$STAGE" "$OUT"
-
-RUNSCM=$STAGE/run.scm
-scripts/boot4-gen-runscm.sh "$ARCH" "$RUNSCM"
-echo "[$BOOT_TAG] generated run.scm: $(wc -l <"$RUNSCM") lines"
+runscm_gen scripts/boot4-gen-runscm.sh "$ARCH"
runscm_scheme1 "$BOOT2/scheme1"
runscm_prelude "$SRC/src/scheme1/prelude.scm"
-runscm_runscm "$RUNSCM"
runscm_input tcc0 "$BOOT3/tcc0"
runscm_input catm "$BOOT2/catm"
-runscm_input_from_src src "tcc-libc/$ARCH/start.S"
-runscm_input_from_src src "tcc-libc/$ARCH/sys_stubs.S"
-runscm_input_from_src src tcc-cc/mem.c
+runscm_input_from_src "tcc-libc/$ARCH/start.S"
+runscm_input_from_src "tcc-libc/$ARCH/sys_stubs.S"
+runscm_input_from_src tcc-cc/mem.c
for f in $LIBTCC1_C_SRCS $LIBTCC1_ASM_SRCS; do
- runscm_input_from_src src "$TCC_LIB_REL/$f"
+ runscm_input_from_src "$TCC_LIB_REL/$f"
done
-runscm_input_from_src src tcc/tcc.flat.c
-runscm_input_from_src src libc/libc.flat.c
-runscm_input_from_src src test-fixtures/boot-hello.c hello.c
+runscm_input_from_src tcc/tcc.flat.c
+runscm_input_from_src libc/libc.flat.c
+runscm_input_from_src test-fixtures/boot-hello.c hello.c
-runscm_export tcc1
-runscm_export tcc2
-runscm_export tcc3
-runscm_export s3-crt1.o
-runscm_export s3-libc.a
-runscm_export s3-libtcc1.a
-runscm_export hello
+runscm_export tcc1 tcc2 tcc3 s3-crt1.o s3-libc.a s3-libtcc1.a hello
runscm_run "${BOOT4_TIMEOUT:-5400}"
# ── fixed-point check (host-side) ─────────────────────────────────────
-# After a codegen-altering tcc patch, tcc1 (built by tcc0 = pre-fix) and
-# tcc2 (built by tcc1 = post-fix codegen but tcc0-emitted binary) won't
-# agree byte-for-byte with tcc3 (built by tcc2 = post-fix throughout):
-# the two-stage rule needs a third bounce to converge. Set
-# TCC_BOOTSTRAP_RELAX_FIXEDPOINT=1 to skip the cmp; the very next boot4
-# run, started from this run's tcc3, will produce tcc2 == tcc3 with no
-# extra knob.
if ! cmp -s "$OUT/tcc2" "$OUT/tcc3"; then
s2=$(wc -c <"$OUT/tcc2")
s3=$(wc -c <"$OUT/tcc3")
diff --git a/scripts/boot5.sh b/scripts/boot5.sh
@@ -8,26 +8,19 @@
## attribute(alias) weak refs, _Complex, x86_64 SSE/x87 inline asm).
##
## ─── Inputs ──────────────────────────────────────────────────────────
-## build/$ARCH/$DRIVER/boot4/tcc3
-## — boot4's verified self-host tcc
-## build/$ARCH/$DRIVER/boot4/libtcc1.a
-## — boot4's tcc runtime archive
-## build/$ARCH/src/src/musl/ — canonical musl tree (overrides merged,
-## deletes applied, alltypes.h/syscall.h
-## generated, per-arch skip filter
-## applied). Built by prep-src.sh +
-## prep-musl.sh.
+## build/$ARCH/$DRIVER/boot4/tcc3 — boot4's verified self-host tcc
+## build/$ARCH/$DRIVER/boot4/libtcc1.a — boot4's tcc runtime archive
+## build/$ARCH/$DRIVER/boot2/{catm, scheme1}
+## build/$ARCH/src/src/musl/ — canonical musl tree (overrides
+## merged, deletes applied,
+## alltypes.h/syscall.h generated,
+## per-arch skip filter applied)
## build/$ARCH/src/src/tcc/stdarg-bridge.h
-## — per-arch __builtin_va_list bridge.
## build/$ARCH/src/src/test-fixtures/boot-hello.c
-## — smoke binary linked at the end.
##
## ─── Tools ────────────────────────────────────────────────────────────
-## In container: scratch + busybox (no libc, no /etc, no resolver).
-## scheme1 evaluates a host-generated run.scm (emitted
-## by scripts/boot5-gen-runscm.sh) against the flat
-## staging root (cwd=/work for podman, cwd=/ for the
-## seed kernel). Same run.scm drives both.
+## scheme1 evaluates a host-generated run.scm (from boot5-gen-runscm.sh)
+## against the flat staging root.
##
## ─── Outputs ─────────────────────────────────────────────────────────
## build/$ARCH/$DRIVER/boot5/libc.a
@@ -35,28 +28,27 @@
## build/$ARCH/$DRIVER/boot5/hello — static, runs in the container
##
## Usage: scripts/boot5.sh <arch>
-## <arch> ∈ {amd64, aarch64, riscv64} for either DRIVER (default podman).
+## <arch> ∈ {aarch64, amd64, riscv64} for either DRIVER (default podman).
set -eu
. scripts/lib-arch.sh
bootlib_init boot5 "${1:-}"
driver_init empty
+require_src
BOOT2=build/$ARCH/$DRIVER/boot2
BOOT4=build/$ARCH/$DRIVER/boot4
SRC=build/$ARCH/src
-OUT=build/$ARCH/$DRIVER/boot5
-STAGE=build/$ARCH/$DRIVER/.boot5-stage
MUSL_DIR=$SRC/src/musl
# ── prerequisites ─────────────────────────────────────────────────────
require_prev "$BOOT4" tcc3
require_prev "$BOOT2" catm scheme1
-[ -e "$BOOT4/libtcc1.a" ] || { echo "[$BOOT_TAG] missing $BOOT4/libtcc1.a (run scripts/boot4.sh $ARCH)" >&2; exit 1; }
-[ -d "$MUSL_DIR" ] || { echo "[$BOOT_TAG] missing $MUSL_DIR — run scripts/prep-src.sh $ARCH and scripts/prep-musl.sh $ARCH" >&2; exit 1; }
-[ -e "$MUSL_DIR/skip.txt" ] || { echo "[$BOOT_TAG] missing $MUSL_DIR/skip.txt — run scripts/prep-musl.sh $ARCH" >&2; exit 1; }
-[ -e "$SRC/src/tcc/stdarg-bridge.h" ] || { echo "[$BOOT_TAG] missing $SRC/src/tcc/stdarg-bridge.h — run scripts/prep-src.sh $ARCH" >&2; exit 1; }
+require_file "$BOOT4/libtcc1.a" "run scripts/boot4.sh $ARCH"
+require_file "$MUSL_DIR" "run scripts/prep-src.sh $ARCH and scripts/prep-musl.sh $ARCH"
+require_file "$MUSL_DIR/skip.txt" "run scripts/prep-musl.sh $ARCH"
+require_file "$SRC/src/tcc/stdarg-bridge.h" "run scripts/prep-src.sh $ARCH"
# ── prepare staging dirs ──────────────────────────────────────────────
# $STAGE/in/ — read-only inputs (becomes /work/in or in/ in tmpfs)
@@ -151,31 +143,24 @@ mkdir -p "$COBJ/crt"
while read -r d; do mkdir -p "$COBJ/$d"; done < "$STAGE/_host/build-objdirs.txt"
# ── generate run.scm and stage chain binaries ─────────────────────────
-RUNSCM=$STAGE/run.scm
-scripts/boot5-gen-runscm.sh "$MUSL_ARCH" "$STAGE/_host" "$RUNSCM"
-echo "[$BOOT_TAG] generated run.scm: $(wc -l <"$RUNSCM") lines, $(wc -c <"$RUNSCM") bytes"
+runscm_gen scripts/boot5-gen-runscm.sh "$MUSL_ARCH" "$STAGE/_host"
runscm_scheme1 "$BOOT2/scheme1"
runscm_prelude "$SRC/src/scheme1/prelude.scm"
-runscm_runscm "$RUNSCM"
# Chain binaries staged at flat in/ root (cwd-relative names in run.scm).
runscm_input tcc "$BOOT4/tcc3"
runscm_input libtcc1.a "$BOOT4/libtcc1.a"
runscm_input catm "$BOOT2/catm"
-runscm_input_from_src src tcc/stdarg-bridge.h tcc-stdarg-bridge.h
-runscm_input_from_src src test-fixtures/boot-hello.c hello.c
+runscm_input_from_src tcc/stdarg-bridge.h tcc-stdarg-bridge.h
+runscm_input_from_src test-fixtures/boot-hello.c hello.c
# Stage the canonical musl tree under in/musl/. Both drivers pick it
# up automatically (podman bind-mounts $STAGE/in; seed packs
# `find in -type f` into the cpio).
-runscm_input_tree_from_src musl src musl
+runscm_input_tree_from_src musl musl
-runscm_export libc.a
-runscm_export crt1.o
-runscm_export crti.o
-runscm_export crtn.o
-runscm_export hello
+runscm_export libc.a crt1.o crti.o crtn.o hello
# boot5 has ~1300 spawns + heavy tcc work; bump qemu memory + timeout for
# the seed driver. Podman ignores QEMU_MEM and uses host memory directly.
diff --git a/scripts/boot6.sh b/scripts/boot6.sh
@@ -1,81 +1,69 @@
#!/bin/sh
-## boot6.sh — build the seed-kernel ELF with boot4's tcc3.
+## boot6.sh — build the seed-kernel ELF/Image with boot4's tcc3.
##
## Drives tcc3 to compile + link the seed kernel directly: no `ld -T
-## kernel.lds`, no objcopy. aarch64 emits the flat Image QEMU expects;
+## kernel.lds`, no objcopy. aarch64 emits the flat Image QEMU expects;
## amd64/riscv64 emit the ELF consumed by QEMU's -kernel path.
##
-## ─── Inputs ──────────────────────────────────────────────────────────
-## build/$ARCH/$DRIVER/boot4/tcc3 — boot4's verified self-host tcc
-## (compiler + linker)
-## build/$ARCH/$DRIVER/boot2/scheme1 — driver runtime
-## build/$ARCH/src/src/kernel/arch/$ARCH/{kernel.S,mmu.c,arch.h}
-## — per-arch boot stub, MMU setup, header
+## ─── Inputs (sources, from canonical tree) ───────────────────────────
+## build/$ARCH/src/src/kernel/arch/$ARCH/{kernel.S, mmu.c, arch.h}
## build/$ARCH/src/src/kernel/kernel.c
-## — DTB parse, MMU bring-up, syscalls,
-## virtio-blk, tmpfs, ELF loader
-## build/$ARCH/src/src/tcc-cc/mem.c memcpy/memset/memmove/memcmp
+## build/$ARCH/src/src/tcc-cc/mem.c
+##
+## ─── Inputs (binaries from prior stages) ──────────────────────────────
+## build/$ARCH/$DRIVER/boot4/tcc3
+## build/$ARCH/$DRIVER/boot2/scheme1
##
## ─── Tools ────────────────────────────────────────────────────────────
-## In container: scratch + busybox (boot2-empty:$ARCH).
-## scheme1 evaluates run.scm against the flat staging
-## root (cwd=/work for podman, cwd=/ for the seed
-## kernel). Same run.scm drives both.
+## scheme1 evaluates a host-generated run.scm (from boot6-gen-runscm.sh)
+## against the flat staging root.
##
## ─── Outputs ─────────────────────────────────────────────────────────
-## build/$ARCH/$DRIVER/boot6/Image — flat arm64 boot Image, byte-format
-## identical in shape to the gcc
-## Makefile's `objcopy -O binary`
-## output. QEMU's `-kernel` detects
-## `ARM\x64` magic at file offset 0x38
-## and follows the arm64 boot
-## protocol, putting DTB phys in x0
-## before jumping to _start.
+## build/$ARCH/$DRIVER/boot6/$KERNEL_NAME
+## aarch64: Image — flat boot Image, byte-format identical to the gcc
+## Makefile's `objcopy -O binary` output. QEMU's `-kernel`
+## detects `ARM\x64` magic at file offset 0x38 and follows
+## the arm64 boot protocol, putting DTB phys in x0 before
+## jumping to _start.
+## amd64/riscv64: kernel.elf — ELF consumed via QEMU's -kernel path.
##
## Usage: scripts/boot6.sh <arch>
-## <arch> ∈ {amd64,aarch64,riscv64} for either DRIVER (default podman).
+## <arch> ∈ {aarch64, amd64, riscv64} for either DRIVER (default podman).
set -eu
. scripts/lib-arch.sh
bootlib_init boot6 "${1:-}"
driver_init empty
+require_src
-OUT_FILE=$KERNEL_NAME
BOOT2=build/$ARCH/$DRIVER/boot2
BOOT4=build/$ARCH/$DRIVER/boot4
SRC=build/$ARCH/src
-OUT=build/$ARCH/$DRIVER/boot6
-STAGE=build/$ARCH/$DRIVER/.boot6-stage
# ── prerequisites ─────────────────────────────────────────────────────
require_prev "$BOOT4" tcc3
require_prev "$BOOT2" scheme1
-[ -d "$SRC" ] || { echo "[$BOOT_TAG] missing $SRC — run scripts/prep-src.sh $ARCH" >&2; exit 1; }
for f in kernel/arch/$ARCH/kernel.S kernel/arch/$ARCH/mmu.c kernel/arch/$ARCH/arch.h kernel/kernel.c tcc-cc/mem.c; do
- [ -f "$SRC/src/$f" ] || { echo "[$BOOT_TAG] missing $SRC/src/$f" >&2; exit 1; }
+ require_file "$SRC/src/$f"
done
# ── stage inputs and run scheme1 + run.scm under $DRIVER ──────────────
. scripts/lib-runscm.sh
runscm_init "$STAGE" "$OUT"
-
-RUNSCM=$STAGE/run.scm
-scripts/boot6-gen-runscm.sh "$ARCH" "$RUNSCM"
-echo "[$BOOT_TAG] generated run.scm: $(wc -l <"$RUNSCM") lines"
+runscm_gen scripts/boot6-gen-runscm.sh "$ARCH"
runscm_scheme1 "$BOOT2/scheme1"
runscm_prelude "$SRC/src/scheme1/prelude.scm"
-runscm_runscm "$RUNSCM"
runscm_input tcc3 "$BOOT4/tcc3"
-runscm_input_from_src src "kernel/arch/$ARCH/kernel.S"
-runscm_input_from_src src kernel/kernel.c
-runscm_input_from_src src "kernel/arch/$ARCH/arch.h"
-runscm_input_from_src src "kernel/arch/$ARCH/mmu.c"
-runscm_input_from_src src tcc-cc/mem.c
+runscm_input_from_src "kernel/arch/$ARCH/kernel.S"
+runscm_input_from_src kernel/kernel.c
+runscm_input_from_src "kernel/arch/$ARCH/arch.h"
+runscm_input_from_src "kernel/arch/$ARCH/mmu.c"
+runscm_input_from_src tcc-cc/mem.c
-runscm_export "$OUT_FILE"
+runscm_export "$KERNEL_NAME"
runscm_run "${BOOT6_TIMEOUT:-1200}"
-echo "[$BOOT_TAG] OK -> $OUT/$OUT_FILE ($(wc -c <"$OUT/$OUT_FILE") bytes)"
+echo "[$BOOT_TAG] OK -> $OUT/$KERNEL_NAME ($(wc -c <"$OUT/$KERNEL_NAME") bytes)"
diff --git a/scripts/lib-arch.sh b/scripts/lib-arch.sh
@@ -5,13 +5,16 @@
#
# bootlib_init <stage> <arch> # validate <arch>, cd to repo root,
# # set ARCH/PLATFORM/KERNEL_NAME/
-# # MUSL_ARCH/DRIVER/BOOT_TAG.
-# driver_init [<image-kind>] # podman: build IMAGE if missing
-# # (image-kind ∈ scratch|empty;
-# # default scratch).
-# # seed: verify boot6 kernel exists.
-# require_prev <dir> <name>... # die helpfully if any <dir>/<name>
-# # is missing or non-executable.
+# # MUSL_ARCH/DRIVER/BOOT_STAGE/BOOT_TAG.
+# driver_init [<image-kind>] # set OUT/STAGE; podman: build IMAGE
+# # if missing (image-kind ∈ scratch|
+# # empty; default scratch); seed:
+# # verify boot6 kernel exists.
+# require_src # die if build/$ARCH/src/ missing.
+# require_prev <dir> <name>... # die if any <dir>/<name> is
+# # missing or non-executable.
+# require_file <path> [<hint>] # die if <path> missing; print a
+# # uniform diagnostic with hint.
#
# After bootlib_init, the following shell vars are set/exported:
# ARCH input architecture token (aarch64|amd64|riscv64)
@@ -21,8 +24,11 @@
# KERNEL_NAME Image (aarch64) | kernel.elf (amd64,riscv64)
# MUSL_ARCH aarch64 | x86_64 | riscv64
# BOOT_TAG "<stage>/<driver>/<arch>" for log prefixes
+# BOOT_STAGE stage name as passed in (boot0|boot1|...)
#
-# After driver_init:
+# After driver_init (boot stages only — prep-* skip it):
+# OUT build/$ARCH/$DRIVER/$BOOT_STAGE (stage output dir)
+# STAGE build/$ARCH/$DRIVER/.$BOOT_STAGE-stage (scratch staging dir)
# podman: IMAGE
# seed: KERNEL_IMAGE, EXTRACT, SEED_ARCH
@@ -41,13 +47,56 @@ bootlib_init() {
podman|seed) ;;
*) echo "[$_stage/$DRIVER/$ARCH] unknown DRIVER=$DRIVER (expected podman|seed)" >&2; exit 2 ;;
esac
+ BOOT_STAGE=$_stage
BOOT_TAG="$_stage/$DRIVER/$ARCH"
+ BOOT_T0=$(date +%s)
case "$ARCH" in
aarch64) PLATFORM=linux/arm64; KERNEL_NAME=Image; MUSL_ARCH=aarch64 ;;
amd64) PLATFORM=linux/amd64; KERNEL_NAME=kernel.elf; MUSL_ARCH=x86_64 ;;
riscv64) PLATFORM=linux/riscv64; KERNEL_NAME=kernel.elf; MUSL_ARCH=riscv64 ;;
esac
- export ARCH ROOT DRIVER PLATFORM KERNEL_NAME MUSL_ARCH BOOT_TAG
+ export ARCH ROOT DRIVER PLATFORM KERNEL_NAME MUSL_ARCH BOOT_TAG BOOT_STAGE BOOT_T0
+ trap _bootlib_finish EXIT
+}
+
+# _bootlib_finish — EXIT trap installed by bootlib_init. Prints
+# `[$BOOT_TAG] done in Xs (cum Ys)` (or `failed after Xs` on error).
+# On success, records the elapsed time so later stages can sum the
+# chain. Cumulative = sum of all per-stage .timing files relevant to
+# the current $ARCH/$DRIVER.
+_bootlib_finish() {
+ _exit=$?
+ [ -n "${BOOT_T0:-}" ] || return 0
+ _elapsed=$(( $(date +%s) - BOOT_T0 ))
+ if [ "$_exit" != 0 ]; then
+ echo "[$BOOT_TAG] failed after ${_elapsed}s (exit=$_exit)" >&2
+ return 0
+ fi
+ # Record this stage's time. Boot stages have OUT (set by driver_init);
+ # the orchestrator (BOOT_STAGE=boot) doesn't write — its time would
+ # double-count. Other stages without OUT (prep-src, prep-musl) write
+ # to a per-arch sidecar dir.
+ if [ "$BOOT_STAGE" != boot ]; then
+ if [ -n "${OUT:-}" ] && [ -d "$OUT" ]; then
+ echo "$_elapsed" > "$OUT/.timing"
+ elif [ -d "build/$ARCH" ]; then
+ mkdir -p "build/$ARCH/.timings"
+ echo "$_elapsed" > "build/$ARCH/.timings/$BOOT_STAGE"
+ fi
+ fi
+ # Cumulative: sum boot-stage timings for this driver + driver-
+ # independent prep timings. Glob may not match — guard each path.
+ _cum=0
+ for _f in \
+ "build/$ARCH/$DRIVER"/*/.timing \
+ "build/$ARCH/.timings"/*
+ do
+ [ -f "$_f" ] || continue
+ _v=$(cat "$_f" 2>/dev/null) || continue
+ case "$_v" in *[!0-9]*|'') continue ;; esac
+ _cum=$((_cum + _v))
+ done
+ echo "[$BOOT_TAG] done in ${_elapsed}s (cum ${_cum}s)"
}
driver_init() {
@@ -56,6 +105,9 @@ driver_init() {
scratch|empty) ;;
*) echo "[$BOOT_TAG] driver_init: image-kind must be scratch|empty (got $_image_kind)" >&2; exit 2 ;;
esac
+ OUT=build/$ARCH/$DRIVER/$BOOT_STAGE
+ STAGE=build/$ARCH/$DRIVER/.$BOOT_STAGE-stage
+ export OUT STAGE
case "$DRIVER" in
podman)
IMAGE=boot2-$_image_kind:$ARCH
@@ -100,3 +152,26 @@ require_prev() {
}
done
}
+
+# require_src — assert build/$ARCH/src/ exists (the canonical generated
+# source tree built by scripts/prep-src.sh). Every bootN.sh needs it.
+require_src() {
+ [ -d "build/$ARCH/src" ] || {
+ echo "[$BOOT_TAG] missing build/$ARCH/src — run scripts/prep-src.sh $ARCH" >&2
+ exit 1
+ }
+}
+
+# require_file <path> [<hint>] — assert <path> exists; print a uniform
+# "[$BOOT_TAG] missing <path> — <hint>" diagnostic on failure.
+require_file() {
+ _path=$1; _hint=${2:-}
+ [ -e "$_path" ] || {
+ if [ -n "$_hint" ]; then
+ echo "[$BOOT_TAG] missing $_path — $_hint" >&2
+ else
+ echo "[$BOOT_TAG] missing $_path" >&2
+ fi
+ exit 1
+ }
+}
diff --git a/scripts/lib-pipeline.sh b/scripts/lib-pipeline.sh
@@ -19,9 +19,10 @@
# DSL (source as `. scripts/lib-pipeline.sh`):
#
# pipeline_init <staging-dir> <out-dir> <driver>
-# pipeline_input <name> <host-path> # may be called repeatedly
-# stage <bin> <argv1> <argv2>... -- <input-names...> -- <output-names...>
-# pipeline_export <name> # may be called repeatedly
+# pipeline_input <name> <host-path> # repeatable
+# pipeline_input_from_src <subpath> [<name>] # from build/$ARCH/src/src/
+# stage <bin> <argv...> -- <inputs...> -- <outputs...>
+# pipeline_export <name>... # one or more
# pipeline_run
#
# `stage` semantics: invoke `<bin>` with argv=[<bin>, <argv1>, ...]; the
@@ -82,15 +83,16 @@ pipeline_input() {
P_INPUT_NAMES="$P_INPUT_NAMES $name"
}
-# pipeline_input_from_src — pull a file from the canonical generated
-# source tree at build/$ARCH/src/{bin,src}/<subpath>. Stages under
-# in/<name> where <name> defaults to basename(subpath); pass an
-# override as the optional third argument when the staged name must
-# differ (e.g. P1.M1 vs P1-aarch64.M1).
+# pipeline_input_from_src <subpath> [<name>]
+# Pull a file from the canonical generated source tree at
+# build/$ARCH/src/src/<subpath>. Stages it under in/<name>;
+# <name> defaults to basename(subpath). For the rare `bin/` case
+# (the seed hex0-seed binary), call pipeline_input directly with
+# build/$ARCH/src/bin/<file>.
pipeline_input_from_src() {
- _kind=$1; _subpath=$2; _name=${3:-}
+ _subpath=$1; _name=${2:-}
[ -n "$_name" ] || _name=$(basename "$_subpath")
- pipeline_input "$_name" "build/$ARCH/src/$_kind/$_subpath"
+ pipeline_input "$_name" "build/$ARCH/src/src/$_subpath"
}
# Look up a token: if it names an input, prefix `in/`; if it names a
@@ -112,6 +114,14 @@ _p_bin_path() {
echo "$b"
}
+# stage <bin> <argv...> -- <inputs...> -- <outputs...>
+#
+# The explicit input/output lists look redundant — most names already
+# appear in <argv...> — but they are not. argv positions are tool-
+# specific: a token like `M0.combined.hex2` is an output of one stage
+# (catm produces it) and an input of the next (hex2 reads it). The
+# framework cannot tell which from the token alone, so each stage
+# declares both lists. Don't try to "simplify" by inferring from argv.
stage() {
bin=$1; shift
P_HEAD_RAW=""; P_IN=""; P_OUT=""; _s=head
@@ -258,7 +268,7 @@ in/$inp"
}
pipeline_export() {
- P_EXPORTS="$P_EXPORTS $1"
+ for _n in "$@"; do P_EXPORTS="$P_EXPORTS $_n"; done
}
pipeline_run() {
diff --git a/scripts/lib-runscm.sh b/scripts/lib-runscm.sh
@@ -18,13 +18,15 @@
# DSL (source as `. scripts/lib-runscm.sh`):
#
# runscm_init <staging-dir> <out-dir>
-# runscm_scheme1 <path> # init=scheme1 (boot2)
-# runscm_prelude <path> # scheme1/prelude.scm
-# runscm_runscm <path> # the host-generated driver
-# runscm_input <name> <host-path> # repeatable; staged at in/<name>
-# runscm_input_tree <prefix> <src-root> # repeatable; tree under in/<prefix>
-# runscm_export <name> # repeatable; out file
-# runscm_run [timeout-s] # default 600s
+# runscm_scheme1 <path> # init=scheme1 (boot2)
+# runscm_prelude <path> # scheme1/prelude.scm
+# runscm_runscm <path> # static driver script
+# runscm_gen <gen-script> <args...> # OR generate run.scm,
+# # log size, register it.
+# runscm_input <name> <host-path> # repeatable; staged at in/<name>
+# runscm_input_tree <prefix> <src-root> # repeatable; tree under in/<prefix>
+# runscm_export <name>... # one or more output names
+# runscm_run [timeout-s] # default 600s
#
# Required env per driver:
# podman: IMAGE, PLATFORM
@@ -50,6 +52,18 @@ runscm_scheme1() { S_SCHEME1=$1; }
runscm_prelude() { S_PRELUDE=$1; }
runscm_runscm() { S_RUNSCM=$1; }
+# runscm_gen <gen-script> <args...>
+# Run a host-side generator that emits run.scm to $S_STAGE_DIR/run.scm,
+# log its size, and register it as the driver script. Used by
+# boot4/5/6 which build their run.scm dynamically.
+runscm_gen() {
+ _gen=$1; shift
+ _runscm=$S_STAGE_DIR/run.scm
+ "$_gen" "$@" "$_runscm"
+ echo "[$BOOT_TAG] generated run.scm: $(wc -l <"$_runscm") lines, $(wc -c <"$_runscm") bytes"
+ S_RUNSCM=$_runscm
+}
+
runscm_input() {
name=$1; src=$2
case "$name" in
@@ -70,24 +84,27 @@ runscm_input_tree() {
done
}
-# runscm_input_from_src — pull a file from the canonical generated
-# source tree at build/$ARCH/src/{bin,src}/<subpath>. Stages under
-# in/<name> where <name> defaults to basename(subpath).
+# runscm_input_from_src <subpath> [<name>]
+# Pull a file from the canonical generated source tree at
+# build/$ARCH/src/src/<subpath>. Stages it under in/<name>;
+# <name> defaults to basename(subpath). For the rare `bin/` case,
+# call runscm_input directly with build/$ARCH/src/bin/<file>.
runscm_input_from_src() {
- _kind=$1; _subpath=$2; _name=${3:-}
+ _subpath=$1; _name=${2:-}
[ -n "$_name" ] || _name=$(basename "$_subpath")
- runscm_input "$_name" "build/$ARCH/src/$_kind/$_subpath"
+ runscm_input "$_name" "build/$ARCH/src/src/$_subpath"
}
-# runscm_input_tree_from_src — same as runscm_input_tree, but the
-# source root is build/$ARCH/src/{bin,src}/<subpath>.
+# runscm_input_tree_from_src <prefix> <subpath>
+# Same as runscm_input_tree, but the source root is
+# build/$ARCH/src/src/<subpath>.
runscm_input_tree_from_src() {
- _prefix=$1; _kind=$2; _subpath=$3
- runscm_input_tree "$_prefix" "build/$ARCH/src/$_kind/$_subpath"
+ _prefix=$1; _subpath=$2
+ runscm_input_tree "$_prefix" "build/$ARCH/src/src/$_subpath"
}
runscm_export() {
- S_EXPORTS="$S_EXPORTS $1"
+ for _n in "$@"; do S_EXPORTS="$S_EXPORTS $_n"; done
}
runscm_run() {
