The goal is to insert device drivers in between.

That makes it easier to write drivers that can be xcompiled.

On the emulator of course...

Gave more priority to CPU in the emulator because initialization would
otherwise be horrendously slow.

Also, made LIST properly handle full 64-chars lines not ending with
a null or CR. Previously, such a line would be accompanied by its
following line, duplicating that line's output.

This allows us to make DO..LOOP more compact.

This was caused by the removal of signed-ness a long while ago.

I'd like to move this code, along with all recipe contents, to blk/.
It's easier if "emul-related" code is gathered at the same place.

Lines edited with the Block editor end with 0xd and (print) spits
spurious CR when printing those lines.

Also, properly initialize ACC in BROWSE.

And use them in all places they make sense.

With all this recent movements, we had a bit of a fragmentation
issue.

This allows us to get rid of the (parse*) indirection.

Not worth the complexity and space.

2 was overkill

I don't believe this, one hour of my life I'll never get back.

This would allow things like temporary giving control to the *CL
line on the TRS-80. For example... A very far fetched example. Not
at all the only *raison d'etre* of the layer...

Also, add word "C!+" and fix PSP leak in FILL.

XPACKed source needing to be kept in bootstrap bin didn't make
much sense. With the removal of stripfc, it makes more sense to keep
the compiled form of stage 1 in repository as a bootstrap binary.

Also, rename it to forth.bin.

This allows us to write LATEST at proper offset in emul/stage1.fs in
cases where we set a nonzero BIN(.

Will be useful as kinds of "#ifdef"s during xcomp.

ref #97

Will be useful for assembling binaries for the TRS-80 which can't start
at addr 0.

Big bootstrapping change which greatly reduces the role of the modern
operating system in the bootstrapping process.

More reliable.

It's now a simple flag updated by the C< implementation. No more
routine pointer.

The goal is to avoid tons of "ok" during a LOAD operation.

This will allow us to read from blkfs in tests.

Less noisy this way

This is done by adding a "C<* override" layer that is reset to 0 on
ABORT.

The problem was that when ABORT happened during a LOAD, we had a
resetted RSP and started from a clean INTERPRET, but LOAD didn't
have the opportunity to restore C<*, which caused it to continue
interpreting from the faulty BLK.

With a C<* override, we don't need to *restore* C<*, we just need to
clear the override.

This is more than cosmetic, it's also highly usable. The presence
or absence of the "ok" message allows us to know whether the command
aborted. Previously, the "> " prompt appeared when the system expected
a prompt in the INTERPRET context, whether the previous command aborted
or not.

Also, this allows us to get rid of that ugly FLAGS global variable.

The way is clear for complete stage1 bootstrapping on the RC2014
target!