memlnaut-nisps/scripts/firmware-common.sh

135 lines
4.2 KiB
Bash
Raw Permalink Normal View History

2026-04-15 17:32:32 +02:00
#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
REPO_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
# PlatformIO project directory (contains platformio.ini). One [env:<alias>]
# per firmware variant defined there — that file IS the variant registry,
# there is no separate in-source list to keep in sync.
FIRMWARE_PROJECT_DIR="${MEMLNAUT_FIRMWARE_PROJECT_DIR:-$REPO_ROOT/firmware/MEMLNaut-NISPS}"
PLATFORMIO_INI="$FIRMWARE_PROJECT_DIR/platformio.ini"
2026-04-15 17:32:32 +02:00
FIRMWARE_VARIANTS=()
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
ACTIVE_FIRMWARE_VARIANT="" # platformio.ini's [platformio] default_envs, if set
2026-04-15 17:32:32 +02:00
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
ensure_pio() {
if command -v pio >/dev/null 2>&1; then
2026-04-15 17:52:39 +02:00
return 0
fi
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
echo "error: pio (PlatformIO Core) is not installed or not on PATH" >&2
echo " This machine uses nix. Run commands through:" >&2
echo " nix-shell -p platformio-core --run '<command>'" >&2
echo " (NOT 'nix-shell -p platformio' — that package wraps the CLI in a" >&2
echo " bubblewrap FHS sandbox that needs a real user namespace and fails" >&2
echo " under containerised/sandboxed shells. platformio-core is the same" >&2
echo " CLI, unwrapped, and works everywhere.)" >&2
exit 1
2026-04-15 17:52:39 +02:00
}
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
# Runs `pio` against the firmware project dir without requiring the caller's
# cwd to be there. Always goes through nix-shell so callers don't need pio
# pre-installed on PATH.
run_pio() {
ensure_pio
pio "$@" -d "$FIRMWARE_PROJECT_DIR"
2026-04-15 17:52:39 +02:00
}
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
load_firmware_variants() {
if [[ ! -f "$PLATFORMIO_INI" ]]; then
echo "error: platformio.ini not found: $PLATFORMIO_INI" >&2
2026-04-15 17:32:32 +02:00
exit 1
fi
mapfile -t FIRMWARE_VARIANTS < <(
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
grep -E '^\[env:[A-Za-z0-9_]+\]' "$PLATFORMIO_INI" | sed -E 's/^\[env:([A-Za-z0-9_]+)\]$/\1/'
)
if [[ ${#FIRMWARE_VARIANTS[@]} -eq 0 ]]; then
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
echo "error: no [env:<variant>] sections found in $PLATFORMIO_INI" >&2
exit 1
fi
ACTIVE_FIRMWARE_VARIANT="$(
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
grep -E '^\s*default_envs\s*=' "$PLATFORMIO_INI" | head -n1 \
| sed -E 's/^\s*default_envs\s*=\s*//' | tr -d '[:space:]'
)"
}
resolve_firmware_variant() {
local requested="$1"
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
local requested_lower variant
requested_lower="$(printf '%s' "$requested" | tr '[:upper:]' '[:lower:]')"
for variant in "${FIRMWARE_VARIANTS[@]}"; do
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
if [[ "$requested_lower" == "$(printf '%s' "$variant" | tr '[:upper:]' '[:lower:]')" ]]; then
printf '%s\n' "$variant"
return 0
fi
done
return 1
}
choose_firmware_variant() {
local requested="${1:-}"
local resolved=""
local variant
local choice
local tty_fd=""
load_firmware_variants
if [[ -n "$requested" ]]; then
if resolved="$(resolve_firmware_variant "$requested")"; then
printf '%s\n' "$resolved"
return 0
fi
echo "error: unknown firmware variant: $requested" >&2
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
echo "available variants (from $PLATFORMIO_INI):" >&2
for variant in "${FIRMWARE_VARIANTS[@]}"; do
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
echo " - $variant" >&2
done
exit 1
fi
if exec {tty_fd}<>/dev/tty 2>/dev/null; then
printf 'Select a firmware variant to build:\n' >&"$tty_fd"
local idx=1
for variant in "${FIRMWARE_VARIANTS[@]}"; do
if [[ "$variant" == "$ACTIVE_FIRMWARE_VARIANT" ]]; then
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
printf ' %d) %s (default)\n' "$idx" "$variant" >&"$tty_fd"
else
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
printf ' %d) %s\n' "$idx" "$variant" >&"$tty_fd"
fi
idx=$((idx + 1))
done
while true; do
printf 'Variant number or name: ' >&"$tty_fd"
IFS= read -r -u "$tty_fd" choice
if [[ "$choice" =~ ^[0-9]+$ ]]; then
if (( choice >= 1 && choice <= ${#FIRMWARE_VARIANTS[@]} )); then
exec {tty_fd}>&-
printf '%s\n' "${FIRMWARE_VARIANTS[choice-1]}"
return 0
fi
elif resolved="$(resolve_firmware_variant "$choice" 2>/dev/null)"; then
exec {tty_fd}>&-
printf '%s\n' "$resolved"
return 0
fi
printf 'Invalid selection. Enter a number from the list or a variant name.\n' >&"$tty_fd"
done
fi
if [[ -n "$ACTIVE_FIRMWARE_VARIANT" ]]; then
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
echo "No variant specified and no interactive terminal available; using platformio.ini's default_envs: $ACTIVE_FIRMWARE_VARIANT" >&2
printf '%s\n' "$ACTIVE_FIRMWARE_VARIANT"
return 0
fi
build(firmware): migrate to PlatformIO and vendor memllib (plan §5) One cut, no dual path. Closes ALIGNMENT defect 3 ("Arduino-CLI build machinery is actively hostile") and vision bullet 4. platformio.ini carries 16 [env:], one per variant, each passing -DMEMLNAUT_MODE_TYPE; selftest passes -DNISPS_SELFTEST=1 instead. The env list IS the registry now — the .ino comment-registry and the NISPS_ST_* token-paste table are deleted rather than migrated. L12 noted that table was already silently missing the currently-shipped SLPWorkshop variant, which is the whole argument against having a second list. Also deleted: the Python/sed machinery that rewrote the COMMITTED .ino on every build, the sketch symlink forest, the global TFT_eSPI User_Setup.h mutation (now -D flags — TFT_eSPI's own documented PlatformIO recipe), the UF2 boot-mount detection stack (upload_protocol=picotool talks to the bootloader directly), and build-firmware-arch.sh entirely. Scripts 683 -> 435 lines. memllib is vendored at lib/memllib/ from upstream e291192; no submodules remain. VENDORED.md records provenance and the re-sync procedure. S9: a firmware-build CI job compiles three representative envs against a cached toolchain and reports per-variant flash/RAM. Firmware is in an automated gate for the FIRST time. The old ci.yml comment justified excluding it as "low verification value" — an assessment that did not survive contact, since the SelfTest variant sat broken for an unknown period calling a DisplayDriver method that did not exist at the pinned memllib commit, and nothing noticed because nothing built it. Verified: all 16 envs build from an empty cache, each within ~520 bytes of the arduino-cli binary it replaces, flash and RAM. Measured as .text+.rodata / .data+.bss+vector+uninitialized — NOT PlatformIO's console line, which double-counts .data on this board. This does not prove the hardware boots; no flash+smoke test was possible and that stays an operator chokepoint. slpworkshop 248232/145028 pafsynth 256880/149716 selftest 216228/17960 (all 16 in the CI log format; none exceeds 2% of a 16 MB flash) Two traps recorded so nobody rediscovers them: vendoring memllib's subdirs without a src/ wrapper makes PlatformIO's library builder silently compile NOTHING while still linking; and project build_flags land BEFORE the framework's own -std=gnu++17 -Os, so build_unflags is required. CORRECTION carried in this commit: the firmware sizes in c19d846's message and the first version of the memllib recon doc were wrong — SLPWorkshop 145348, PAFSynth 145300, SelfTest 141840. They came from building variants in sequence through a SHARED incremental arduino-cli build directory, which reused stale objects and under-reported by ~75 KB. Clean-cache rebuilds of the identical commit give 216736/18492 for SelfTest. The real cost of the memllib upstream bump is +216 bytes flash, not +316. Never measure firmware size through a reused build dir. HISTORY NOTE: this commit and the docs commit before it were rebuilt (force-push, 2026-07-21) so that each contains only what its message describes. The first versions had the firmware deletions stranded in the docs commit by a shared-index race between concurrent agents; content is byte-identical to the originals. Gates: run-all-tests.sh ALL GREEN (nisps/ untouched by this change beyond include paths); 16/16 pio envs build.
2026-07-21 20:17:58 +02:00
echo "error: no variant specified, no interactive terminal, and no default_envs set in $PLATFORMIO_INI" >&2
exit 1
}