After successful refill instructions, sample the Thumb IWRAM target and target+2 halfwords to establish bus history. Share ARM/Thumb refill classification with nominal timing. Respect restored instruction state, completed data accesses, region limits, and diagnostic rollback; keep instruction execution and nominal costs unchanged.
Add original regressions for ARM/Thumb BX, Thumb branches and BL, PC writes, stack/block PC loads, status-restoring returns, real SWI/IRQ returns, both target alignments, mirrors/wrap, captured sample ordering, load lanes, failure retention, BIOS isolation, and DMA invalidation. Document official ARM7TDMI refill evidence and the distinction from a full pipeline.
Darwin arm64 validation: 917 workspace tests and 848 core/demo tests per debug/release profile; 39 preparation tests; native ROM windows in both profiles; ten graphics/window smoke modes; fmt, clippy, and rustdoc. Public ARM, Thumb, memory, and BIOS reports remain identical to prior passing results and between profiles. Cold startup, ARM-target refill history, and DMA-to-CPU ordering remain incomplete; no hardware-conformance or commercial-game claim.
Track a separate IWRAM latch with known lanes and transactional updates. Sequential Thumb prefetches and IWRAM byte/halfword/word reads and writes drive only addressed lanes. Other regions and host inspection do not replace the latch. Failed instructions discard staged changes.
Keep cold, refill, discontinuous-entry, IRQ, and DMA-transition history conservative. Share Thumb refill classification with nominal timing, including PC writes to fallthrough. Successful DMA and accepted machine IRQs invalidate history; failed DMA preserves it. Document source evidence, bounded scope, and remaining hardware-validation limits.
Add original lane, alignment, mirror/wrap, signed/unaligned-load, mode, block-transfer, rollback, isolation, control-flow, IRQ/DMA, and nominal-cost regressions. Darwin arm64 validation: 905 workspace tests and 836 core/demo tests per debug/release profile; 39 preparation tests; native ROM windows in both profiles; ten graphics/window smoke modes; fmt, clippy, and rustdoc. Public ARM, Thumb, memory, and BIOS reports remain identical to prior passing reports and between profiles.
Repeat the mapped PC+4 halfword for Thumb code in EWRAM, palette RAM, VRAM, and ROM. Use the aligned PC+4 word for BIOS/OAM. Keep retained BIOS history separate from current unused-memory reads. Reuse one instruction snapshot without extra data cycles or host/fetch/DMA fallbacks.
Keep IWRAM and 16 MiB region-crossing fetches diagnostic until prior bus history is modeled. Add original regressions for regions, alignments, mirrors, physical wrap, signed and unaligned loads, block/stack/PC loads, aliases, modes, short ROMs, strict errors, BIOS isolation, and nominal device progression. Record documented rules and remaining limits.
Darwin arm64 validation: 890 workspace tests and 821 core/demo tests in each of debug and release; 39 preparation tests; native ROM windows in both profiles; ten window/graphics smoke modes; fmt, clippy, and rustdoc. Public ARM, Thumb, memory, and BIOS reports remain identical to previous passing results and between profiles. No physical-hardware or independent public Thumb open-bus conformance claim.
Snapshot the full word at (executing PC+4)&~3 for Thumb BIOS execution, using strict image reads and incoming instruction state. Keep ARM PC+8 open bus separate and unchanged. Commit snapshots only after successful steps; preserve diagnostic behavior for unavailable lookahead.
Add original regression coverage for both code alignments, load widths and lanes, signed and unaligned values, processor modes, sequential fetches, ARM/Thumb transitions, SWI, RAM-backed POP exits, BIOS boundaries, failure retention, and nominal device progression. Record source evidence and explicit pipeline/hardware-validation limits.
Darwin arm64 validation: 878 workspace tests and 809 core/demo tests in each of debug and release; 39 Python tests; native ROM window tests in both profiles; ten window/graphics smoke modes; fmt, clippy, and rustdoc. Public ARM, Thumb, memory, and BIOS reports match prior passing reports and match between profiles. No physical-hardware validation claimed.
Retain the supplied image PC+8 snapshot after successful ARM BIOS instructions. CPU data reads from outside BIOS use its byte lanes; reads inside BIOS and host inspection remain mapped reads. Keep nominal timing and diagnostic atomicity. Unknown history, Thumb BIOS execution, and missing lookahead remain explicit limits rather than guessed values.
Add original regressions for ARM/Thumb callers, all CPU modes, widths and rotation, synthetic SWI/IRQ returns, original boot provenance, context isolation, history invalidation, and device progress. No Nintendo firmware or opcode constants are used to synthesize results.
Darwin arm64 validation: 861 workspace tests and 792 core/demo tests in debug and release; 39 Python tests; native ROM windows; all ten demo smoke tests; fmt, clippy, rustdoc, and docs links. Public ARM, Thumb, and memory ROMs still pass. The public BIOS ROM still fails firmware-specific test 1, now reading its actual replacement-BIOS prefetch rather than raw vector bytes.
Correct the shared Thumb block-store PC offset from four to six bytes. Retain ARM PC+12, 64-byte empty-list writeback, normal load behavior, alignment, flags, and nominal timing. Correct prior internal expectations and add original regressions for bases, banks, alignment, ROM windows, writeback, clocks, and atomic errors.
The pinned public Thumb ROM now passes its unchanged ARM-state r7 checkpoint after 807 steps and 4170 nominal cycles. The pinned ARM ROM still passes. Neither external ROM nor result assertion is modified.
Validated on Darwin arm64: 848 workspace tests in debug and release; 779 core/demo tests separately; 21 Python adapter tests; both public suites in both profiles; native ROM-window tests; all ten demo window smoke tests; fmt, clippy, rustdoc, and local documentation links.
Snapshot mapped PC+8 bytes for ARM data reads in the documented unused ranges. Preserve strict host/fetch/Thumb/DMA diagnostics, missing-ROM errors, existing write rules, and nominal timing. Add original regressions and document the bounded model.
The unmodified pinned jsmolka ARM ROM now reaches its verified checkpoint with r12=0 in debug and release: 1344 steps and 11257 nominal cycles. No ROM, manifest, or assertion changes.
Validated on Darwin arm64: 842 workspace tests in debug and release; 773 core/demo tests separately; Python adapter tests; native ROM windows; all ten demo window smoke tests; fmt, clippy, and rustdoc.
Allow ARM LDR/LDRB/LDRH/LDRSB/LDRSH to share non-PC base and destination registers with writeback. Keep the loaded value after computing addresses from incoming operands. Preserve PC restrictions, unsupported user transfers, store behavior, and atomic memory diagnostics.
Add original regressions for index modes, offsets, widths, alignment, sign extension, all active banks, timing, I/O/IRQ ordering, and failures. The unchanged pinned public ARM ROM progresses past tests 360 and 361, then fails at test 362 on an unmapped 0x80000000 read; no full-suite pass is claimed.
Verified on Darwin arm64: debug/release workspace and core suites, Python adapter tests, Clippy, rustdoc, native ROM-window checks, and all ten demo smoke tests. Document the new behavior and next memory-access blocker.
Restore CPSR from SPSR for ARM TST/TEQ/CMP/CMN with unused Rd=15 when the current mode has an SPSR. Preserve sequential PC and nominal timing; use normal ALU flags in User/System. Keep ordinary result-writing PC return validation unchanged.
Reproduced independent jsmolka/gba-tests ARM failure 234 before this fix. Original regressions cover flags, masks, bank restoration, operand forms, conditions, timing, and atomic errors. The public ROM progresses to a separate unsupported load/writeback alias at test 360; no full public-suite pass is claimed.
Move the README's hardware notes, demo controls, environment setup, test
notes, file map, and references into docs/ and docs/hardware/. The README
is now an overview with a docs index. AGENTS.md gives coding agents a short
map of crates, commands, and where to read and write documentation.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
