Previously we've wrongly estimated that the vector with opcodes has the same
length as the number of words, but:
CALLW takes an argument denoting the index of the function on data stack
PUSHQ takes an argument denoting the index of the literal on data stack
Moreover rename some dictionary entries for more descriptive names and add a new
word .V that prints values stored in the environment.
In the first sketch they operated on stack frames, but this approach was flawed
because a) leaving things on the stack above frame prevented it from growing
leading to an internal error, b) closing the frame wiped all values deposited on
the stack that were above the frame base.
This commit takes a correction over things corrected (and rebased) in
refactor-stacks branch.
consturctors nucl_stack_to_foo remove elements from the stack and leave the
parsed element on the stack.
Move specials at the beginning.
CALLW calls a word from the data stack. The word differs from normal functions
in that it takes no arguments (so the call does not modify the data stack).
To allow words using the stack across calls (like in "real" forth) don't unwind
the stack on exit if there are no locals.
ecl_alloc_object, ecl_free_object
ecl_alloc, ecl_alloc_manual, ecl_alloc_atomic, ecl_dealloc
Moreover move all top-level ops to memory.d so they are not reliant on mem_gc.
The stubbed allocator uses manual memory managament for all ops.
[wip] because we should adjust ecl_make_stack too
MSVC does not allow for specifying /std:c99 so we need c11. We don't rely on the
default standard because it does not allow for static struct initializers.
This table contains symbols that are essential to the core runtime: ECL_T,
ECL_UNBOUND, ECL_SIGNAL_HANDLERS, ECL_RESTART_CLUSTERs, ECL_INTERRUPTS_ENABLED,
ECL_ALLOW_OTHER_KEYS and ECL_UNBOUND.
The table is initialized with constexpr, so it is possible to use its elements
in static elements. We also add ecl_def_function to ecl-inl to allow
appropriating C functions into Lisp world at top level.
Previously we've opencoded calls to these functions, although they may be nicely
abstracted with static inline functions. This change improves code readibility
and portability.
We also remove conditionalization for garbage collector inclusion in autotools.
When we propose an alternative gc, then we may decide to put them back, or to
add necessary ifdef statements directly in files.
Moreover untangle c-stack from the gc code and assign the stack base with a
rough guess only when it is not initialized yet (GC will always fill it).
Finally remove a kludge from ecl_adopt_cpu and disable colleciton until the cpu
is fully initialized.
I've also renamed *HANDLER-CLUSTERS* to a more appropriate *SIGNAL-HANDLERS*.
Currently this symbol is imported to the SYSTEM package, although this may be
revised in the future to cater to multiple global environments. Alternatively
the SYSTEM package may be common to all runtimes.
Instead of storing lists in *HANDLER-CLUSTERS*, we define functions that are
called unconditionally on the handler. HANDLER-BIND defines that function to be
a typecase that is dispatched based on the conditiont type, as specified by CL.
This change will aid further refactor.
Both tags have a special meaning to the low-level runtime (most notably the
frame stack). Extracting them from "all symbols" is a step towards multiple
runtimes.
We include captured functions, blocks and variables along with local
variables. This fixes#799.
Moreover DECODE-IHS-ENV is deperacated and more DWIM:
- calls DECODE-IHS-LOCALS for old arguments
- appends DECODE-IHS-LOCALS and DECODE-IHS-LEXENV for ihs index
DECODE-IHS-LOCALS and DECODE-IHS-LEXENV are responsible for decoding appropriate
environments.
We deprecate the function si_ihs_env in favor of more explicit si_ihs_lex and
si_ihs_lcl, but the former is left for backward compatibility with SLIME/SLYNK
because they call it to query the environment to add locals to the backtrace.
Since ~recently we store local variables in the bytevm on the stack. Also, the
native comipler under specified debug options, stores locals in ihs, but it has
nothing to do with the lexical environment. So it feels justified to push both
to a separate field.
It may seem like this proclamation is invalid sincd !346, but the divergence
happens much earlier. 8c0314022c introduces a
feature where c-compiled code can also add debug information, and in that case
the environment is a vector, so the proclamation back then should be:
(proclamation si:ihs-env (si::index) (or list vector))
Later when we've changed the representation, it should be changed to
(proclamation si:ihs-env (si::index) (or null vector))
Where NULL denotes "no lexical environment".
2025-11-28 11:57:09 +01:00
12 changed files with 145 additions and 88 deletions
