docs(v4.0.0): MACHINE.md -- the ROM is whatever size it takes; not a question
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
co-authored by
Claude Opus 5.5
parent
e429d30a7e
commit
3eca84e433
@@ -42,6 +42,8 @@ vocabularies. These are `.f18` source.
|
||||
bitstream is. That build is the only thing done off the node. It produces one ROM per cell
|
||||
width (32 and 64 bits).
|
||||
|
||||
**Size.** Whatever it takes.
|
||||
|
||||
**Read-only.** A store into the ROM is an address fault. Seal and `COLD` never go below it.
|
||||
|
||||
## 3. The boot
|
||||
@@ -129,12 +131,10 @@ Each step is accepted as `.claude/CLAUDE.md` says:
|
||||
1080 lines over 64 characters. Proposed: `.f18` is plain text with no block format.
|
||||
3. **The `v3/` tree.** Deleted, or kept out of every build but left in the repository? Kept,
|
||||
it lets `mkpost.py` write POST's cases again.
|
||||
4. **ROM size.** It is whatever the assembler and blob reader take. It is set when step 1 is
|
||||
built.
|
||||
5. **Where `starforth.4th` loads.** Proposed: with `forth79.4th`, by the kernel and before
|
||||
4. **Where `starforth.4th` loads.** Proposed: with `forth79.4th`, by the kernel and before
|
||||
POST, because v3's StarForth words are system words and v3's POST covers them. Layer 3 is
|
||||
`init.4th`'s.
|
||||
6. **Where the standard's names live.** "100 percent" in `forth79.4th`: does it name all 130
|
||||
5. **Where the standard's names live.** "100 percent" in `forth79.4th`: does it name all 130
|
||||
Required words, including the ones that are single opcodes (`DUP`, `+`, `@`)? Proposed: yes.
|
||||
The `.f18` primitives carry their own names, and `forth79.4th` gives every standard word its
|
||||
standard name and behaviour. The `.f18` layer then holds no FORTH-79 names at all.
|
||||
|
||||
Reference in New Issue
Block a user