* libguile/goops.h (SCM_MCACHE_N_SPECIALIZED)
(SCM_SET_MCACHE_N_SPECIALIZED, SCM_INITIAL_MCACHE_SIZE)
(scm_make_method_cache, scm_memoize_method, scm_mcache_lookup_cmethod)
(scm_mcache_compute_cmethod):
* libguile/goops.c: Remove these procedures which managed the method
cache. There's still a slot there but it's not initialized. The method
cache is no longer necessary.
* module/oop/goops/dispatch.scm (memoize-method!): Change to not take a
"cache" argument.
* libguile/eval.i.c:
* libguile/vm-i-system.c: Remove dispatch via the method cache.
* libguile/eval.i.c (CEVAL, SCM_APPLY): Dispatch applicable structs
before pure generics. In practice what this means is that we never hit
the mcache case, because all pure generics are applicable structs.
We're moving over to having generics dispatch themselves. Also, they
don't prepend the struct as an arg; in order to have that effect, the
user has closures.
* libguile/goops.c (scm_apply_generic, scm_call_generic_0):
(scm_call_generic_1, scm_call_generic_2, scm_call_generic_3): Dispatch
directly to the struct procedures.
(scm_var_make_extended_generic): Remove a duplicate definition for
scm_var_make_extended_generic.
(create_standard_classes): Mark all instances of
<applicable-struct-class> (themselves classes) as applicable classes.
Meaning: generics are now applicable structs.
* libguile/goops.h (SCM_CLASS_CLASS_LAYOUT): The hashsets are actually
uw slots -- or at least, making subclasses maps the int slots to be uw
slots
* libguile/vm-i-system.c (call, goto/args, mv-call): Dispatch applicable
structs in the VM.
* module/oop/goops/dispatch.scm (emit-linear-dispatch): Fix bug in the
non-rest cache miss case.
(delayed-compile): Rework to avoid fluids.
(cache-dispatch): Don't call `equal?', it causes bootstrapping
problems with the primitive-generic equal?. Using our own version is
faster anyway.
I tried to split this one, and I know it's a bit disruptive, but this
stuff really is one big cobweb. So instead we'll pretend like these are
separate commits, by separating the changelog.
Applicable struct runtime support.
* libguile/debug.c (scm_procedure_source):
* libguile/eval.c (scm_trampoline_0, scm_trampoline_1)
(scm_trampoline_2):
* libguile/eval.i.c (CEVAL):
* libguile/goops.c (scm_class_of):
* libguile/procprop.c (scm_i_procedure_arity):
* libguile/procs.c (scm_procedure_p, scm_procedure, scm_setter): Allow
for applicable structs. Whee!
* libguile/deprecated.h (scm_vtable_index_vtable): Define as a synonym
for scm_vtable_index_self.
(scm_vtable_index_printer): Alias scm_vtable_index_instance_printer.
(scm_struct_i_free): Alias scm_vtable_index_instance_finalize.
(scm_struct_i_flags): Alias scm_vtable_index_flags.
(SCM_STRUCTF_FLAGS): Be a -1 mask, we have a whole word now.
(SCM_SET_VTABLE_DESTRUCTOR): Implement by hand.
Hidden slots.
* libguile/struct.c (scm_make_struct_layout): Add support for "hidden"
fields, writable fields that are not visible to make-struct. This
allows us to add fields to vtables and not break existing make-struct
invocations.
(scm_struct_ref, scm_struct_set_x): Always get struct length from the
vtable. Support hidden fields.
* libguile/goops.c (scm_class_hidden, scm_class_protected_hidden): New
slot classes, to correspond to the new vtable slots.
(scm_sys_prep_layout_x): Turn hidden slots into 'h'.
(build_class_class_slots): Reorder the class slots to account for
vtable fields coming out of negative-land, for name as a vtable slot,
and for hidden fields.
(create_standard_classes): Define <hidden-slot> and
<protected-hidden-slot>.
Clean up struct.h.
* libguile/struct.h: Lay things out cleaner. There are no more hidden
(negative) words. Names are nicer. The exposition is nicer. But the
basics are the same. The incompatibilities are that <vtable> has more
slots now, and that scm_alloc_struct's signature has changed. The
former is ameliorated by the "hidden" slots mentioned before, and the
latter, well, it was always a very internal thing...
(scm_t_struct_finalize): New type, a finalizer function to be run when
instances of a vtable are collected.
(scm_t_struct_free): Removed, structs' data is managed by the GC now,
and not freed by vtable functions.
* libguile/struct.c: (scm_vtable_p): Now we keep flags on
vtable-vtables, so this check is cheaper.
(scm_alloc_struct): No hidden words. Yippee.
(struct_finalizer_trampoline): Entersify.
(scm_make_struct): No need to babysit extra words, though now we have
to babysit flags. Propagate the vtable, applicable, and setter flags
appropriately.
(scm_make_vtable_vtable): Update for new simplicity.
(scm_print_struct): A better printer.
(scm_init_struct): Define <applicable-struct-vtable>, a magical vtable
like CL's funcallable-standard-class. Also define
<applicable-struct-with-setter-vtable>.
Remove foreign object implementation.
* libguile/goops.h:
* libguile/goops.c (scm_make_foreign_object, scm_make_class)
(scm_add_slot, scm_wrap_object, scm_wrap_component): Remove, these
were undocumented and unworking.
Clean up goops.h, a little.
* libguile/goops.h:
* libguile/goops.c: Also clean up.
* module/oop/goops/dispatch.scm (hashset-index): Adapt for new hashset
index.
Entities were meant to be a form of applicable struct. Unfortunately,
the implementation is intertwingled with generics. Removing them, for
now, will make it possible to cleanly re-add applicable struct support.
* libguile/struct.h (SCM_STRUCTF_ENTITY): Remove.
(SCM_STRUCTF_GOOPS_HACK): New flag; sigh.
* libguile/struct.c (scm_make_struct): We make "entity" structs if the
GOOPS_HACK flag is set. This will be fixed when we rework flags and
remove hidden words.
* libguile/goops.c (scm_class_of): Structs are not applicable, for now
at least.
(scm_sys_inherit_magic_x, scm_basic_basic_make_class)
(scm_sys_allocate_instance, scm_sys_set_object_setter_x):
(make_struct_class): Adapt for no more entities (and thus no entity
flag).
(create_standard_classes): For some reason, generic functions were
getting the LIGHT flag set, after the ENTITY flag was removed; so for
now explicitly clear that flag.
* libguile/goops.h (SCM_GENERIC_SETTER, SCM_SET_GENERIC_SETTER): New
macros.
* libguile/objects.h:
* libguile/objects.c: Remove code for entities.
* libguile/debug.c: (scm_procedure_source): Only work with generics.
* libguile/eval.c (scm_trampoline_0, scm_trampoline_1)
(scm_trampoline_2): Only handle generics.
* libguile/eval.i.c (CEVAL): #ifdef out the pieces about entities.
* libguile/procprop.c (scm_i_procedure_arity): Remove support for
entities.
* libguile/procs.c (scm_procedure_p, scm_procedure, scm_setter): Remove
entity support.
* libguile/goops.h (SCM_GENERIC_METHOD_CACHE)
(SCM_SET_GENERIC_METHOD_CACHE): Two new macros; the same as
SCM_[SET_]ENTITY_PROCEDURE, but more reflecting the reality of the
generic hack.
* libguile/eval.i.c:
* libguile/goops.c:
* libguile/objects.c:
* libguile/vm-i-system.c: Use the new macros when it is appropriate to
do so.
* Renumbers the IFLAG constants.
* Adds several macros related to boolean type tests, null tests, and
boolean-truth testing (including lisp-style boolean-truth tests).
* Adds compile-time checks to verify the necessary IFLAG numbering
properties needed for the checks to work properly.
* Changes some existing code to use the new optimized macros, without
changing the semantics of the code at all (except that scm_is_bool
is changed from a function to a macro).
I added the following macros, whose names explicitly state how %nil
should be handled. See the comments in the patch for more information
about these.
scm_is_false_assume_not_lisp_nil scm_is_true_assume_not_lisp_nil
scm_is_false_and_not_lisp_nil scm_is_true_or_lisp_nil
scm_is_false_or_lisp_nil scm_is_true_and_not_lisp_nil
scm_is_lisp_false scm_is_lisp_true
scm_is_null_assume_not_lisp_nil
scm_is_null_and_not_lisp_nil
scm_is_null_or_lisp_nil
scm_is_bool_and_not_lisp_nil
scm_is_bool_or_lisp_nil
The following already-existing macros are defined as aliases, such
that their semantics is unchanged (although scm_is_bool used to be a
function and is now a macro).
scm_is_null --> scm_is_null_and_not_lisp_nil
scm_is_false --> scm_is_false_and_not_lisp_nil
scm_is_true --> scm_is_true_or_lisp_nil
scm_is_bool --> scm_is_bool_and_not_lisp_nil
(I still believe that these should be changed to versions that handle
%nil properly, but await approval on that point, so these patches do
not make those changes)
Also, if the preprocessor macro SCM_ENABLE_ELISP is not true (this
macro already existed and was used in lang.h), all overheads
associated with %nil handling are eliminated from the above macros.
* libguile/tags.h (SCM_BOOL_F, SCM_BOOL_T, SCM_UNSPECIFIED)
(SCM_UNDEFINED, SCM_UNBOUND, SCM_ELISP_NIL): Renumber, so that a
number of important distinctions (false versus true, end-of-list, etc)
can be made by masking a single bit. Also define a number of
build-time tests to assert that this condition holds.
* libguile/boolean.h (scm_is_false_and_not_nil, scm_is_true_or_nil)
(scm_is_false_assume_not_nil, scm_is_true_assume_not_nil):
(scm_is_false_or_nil, scm_is_true_and_not_nil)
(scm_is_bool_or_nil, scm_is_bool_and_not_nil): New exciting macros to
test certain boolean/end-of-list properties.
(scm_is_false, scm_is_true): Use a restrictive definition, where only
SCM_BOOL_F is false. Should probably change in the future.
(scm_is_bool): Incompatible change: changed to be a macro. Was a
function before. Probably should allow nil as a boolean, but that will
be for a later patch.
(scm_is_lisp_false, scm_is_lisp_true): New macros, implementing the
standard Lisp boolean predicates, where '() is actually false.
* libguile/eval.i.c (CEVAL): Fix a number of false-or-nil and similar
tests to use the new macros.
* libguile/lang.h (SCM_NULL_OR_NIL_P): Use scm_is_null_or_nil.
* libguile/pairs.c: Add a compile-time check that null and nil differ by
only one bit.
* libguile/pairs.h (scm_is_null_and_not_nil, scm_is_null_assume_not_nil)
(scm_is_null_or_nil): New exciting macros!
(scm_is_null): Just be scm_is_null_and_not_nil, for now.
* libguile/print.c: Adapt to the reordering, and print suitably nasty
things for the not-to-be-used values.
This fixes a bug introduced in e20d7001c3
and reported by Neil.
* libguile/eval.i.c (CEVAL)[DEVAL]: Don't duplicate ARG1 in
`debug.info->a.args' for gsubr stack frames.
(scm_apply): Likewise.
* test-suite/tests/eval.test ("stacks")["arguments of a gsubr stack
frame"]: New test.
The symbol's characters are only accessed in case they are needed
for an error message. This can be avoided by passing the symbol
all the way to a error message function.
* libguile/__scm.h (SCM_WTA_DISPATCH_1_SUBR): new macro
* libguile/error.c (scm_i_wrong_type_arg_symbol): new error function
* libguile/error.h: declaration of scm_i_wrong_type_arg_symbol
* libguile/eval.c (call_dsubr_1): use new macro SCM_WTA_DISPATCH_1_SUBR
to avoid having to unpack the symbol's chars
* libguile/eval.i.c: use new macro SCM_WTA_DISPATCH_1_SUBR
* libguile/tags.h (scm_tc7_program):
* libguile/programs.h: Programs now have their own tc7 code. Fix up the
macros appropriately.
* libguile/programs.c: Remove smobby bits, leaving marking, printing,
and application for other parts of Guile.
* libguile/debug.c (scm_procedure_source):
* libguile/eval.c (scm_trampoline_0, scm_trampoline_1)
(scm_trampoline_2): Add cases for tc7_program.
* libguile/eval.i.c (CEVAL, SCM_APPLY):
* libguile/evalext.c (scm_self_evaluating_p):
* libguile/gc-card.c (scm_i_sweep_card, scm_i_tag_name):
* libguile/gc-mark.c (1):
* libguile/print.c (iprin1):
* libguile/procs.c (scm_procedure_p, scm_thunk_p)
* libguile/vm-i-system.c (make-closure): Adapt to new procedure
representation.
* libguile/procprop.c (scm_i_procedure_arity): Do the right thing for
programs.
* test-suite/tests/procprop.test ("procedure-arity"): Arity test now
succeeds.
* libguile/goops.c (scm_class_of): Programs now belong to the class
<procedure>, not a smob class.
* libguile/vm.h (struct vm, struct vm_cont):
* libguile/vm-engine.c (vm_engine):
* libguile/frames.h (SCM_FRAME_BYTE_CAST, struct vm_frame):
* libguile/frames.c (scm_c_make_vm_frame): Fix usages of scm_byte_t,
changing them to scm_t_uint8.
* libguile/eval.i.c (CEVAL): Update calls to `scm_i_gsubr_apply ()' with
a fixed number of arguments. Use `scm_i_gsubr_apply_list ()' for
calls with a list of arguments of unknown length.
(SCM_APPLY): Use `scm_i_gsubr_apply_list ()' instead of
`scm_i_gsubr_apply ()'.
* libguile/gsubr.c (gsubr_apply_raw): New.
(scm_i_gsubr_apply): Change to take a C vararg list instead of a
Scheme list. Use `gsubr_apply_raw ()'.
(scm_i_gsubr_apply_list): Use `gsubr_apply_raw ()'.
* libguile/gsubr.h (scm_i_gsubr_apply): Update prototype.
(scm_i_gsubr_apply_list): New declaration.
The idea is to introduce `gsubrs' whose arity is encoded in their type
(more precisely in the sizeof (void *) - 8 MSBs). This removes the
indirection introduced by cclos and simplifies the code.
* libguile/__scm.h (CCLO): Remove.
* libguile/debug.c (scm_procedure_source, scm_procedure_environment):
Remove references to `scm_tc7_cclo'.
* libguile/eval.c (scm_trampoline_0, scm_trampoline_1,
scm_trampoline_2): Replace `scm_tc7_cclo' with `scm_tc7_gsubr'.
* libguile/eval.i.c (CEVAL): Likewise. No longer make PROC the first
argument. Directly invoke `scm_gsubr_apply ()' instead of jump to the
`evap(N+1)' label or call to `SCM_APPLY ()'.
* libguile/evalext.c (scm_self_evaluating_p): Remove reference to
`scm_tc7_cclo'.
* libguile/gc-card.c (scm_i_sweep_card, scm_i_tag_name): Likewise.
* libguile/gc-mark.c (scm_gc_mark_dependencies): Likewise.
* libguile/goops.c (scm_class_of): Likewise.
* libguile/print.c (iprin1): Likewise.
* libguile/gsubr.c (create_gsubr): Use `unsigned int's for REQ, OPT and
RST. Use `scm_tc7_gsubr' instead of `scm_makcclo ()' in the default
case.
(scm_gsubr_apply): Remove calls to `SCM_GSUBR_PROC ()'.
(scm_f_gsubr_apply): Remove.
* libguile/gsubr.h (SCM_GSUBR_TYPE): New definition.
(SCM_GSUBR_MAX): Changed to 33.
(SCM_SET_GSUBR_TYPE, SCM_GSUBR_PROC, SCM_SET_GSUBR_PROC,
scm_f_gsubr_apply): Remove.
* libguile/procprop.c (scm_i_procedure_arity): Remove reference to
`scm_tc7_cclo'; add proper handling of `scm_tc7_gsubr'.
* libguile/procs.c (scm_makcclo, scm_make_cclo): Remove.
(scm_procedure_p): Remove reference to `scm_tc7_cclo'.
(scm_thunk_p): Likewise, plus add proper `scm_tc7_gsubr' handling.
* libguile/procs.h (SCM_CCLO_LENGTH, SCM_MAKE_CCLO_TAG,
SCM_SET_CCLO_LENGTH, SCM_CCLO_BASE, SCM_SET_CCLO_BASE, SCM_CCLO_REF,
SCM_CCLO_SET, SCM_CCLO_SUBR, SCM_SET_CCLO_SUBR, scm_makcclo,
scm_make_cclo): Remove.
* libguile/stacks.c (read_frames): Remove reference to `scm_f_gsubr_apply'.
* libguile/tags.h (scm_tc7_cclo): Remove.
(scm_tc7_gsubr): New.
(scm_tcs_subrs): Add `scm_tc7_gsubr'.
* libguile/gsubr.c (scm_gsubr_apply): Make SELF the first argument
instead of the first element of ARGS.
* libguile/gsubr.h: Update.
* libguile/eval.i.c (CEVAL): Update.
* libguile/eval.i.c (type_dispatch, apply_vm_cmethod)
(apply_memoized_cmethod): Tweak the nastiness a bit more so as to deal
with the '(no-method) empty entries. I would like to stop the search if
the cdr isn't a pair, but currently with the inlined memoized bits, the
cdr is a pair. The fix would be to make the memoizer return a procedure
and not the already-inlined bits -- slightly slower but the vm will be
faster anyway.
* libguile/objects.c (scm_mcache_lookup_cmethod): Same fixes here.
* oop/goops/dispatch.scm (cache-hashval, cache-try-hash!): Allow non-list
cmethod tails.
* ice-9/boot-9.scm (make-modules-in): Change to make sure that we are
making modules in modules; that is, that a global binding of `compile'
doesn't prevent a module from importing a submodule named `compile'.
(resolve-module): Clean up a bit, and serialize the logic.
* libguile/objects.c (scm_mcache_lookup_cmethod, scm_apply_generic):
* libguile/eval.i.c (CEVAL): Now that cmethod entries can have a program
as their tail instead of a memoized proc, we have to change the halting
condition on the method cache search, in both places: the one that's
inlined into eval.i.c and the one in objects.c. If the cmethod isn't a
pair, apply it.
* libguile/goops.c (make): In the `make' procedure that's used before
GOOPS is booted, bind #:formals, #:body, and #:compile-env on methods.
* oop/goops/compile.scm (compute-entry-with-cmethod): There was a
terrible trick here that involved putting a dummy pair in the cache,
then modifying it in place with the result of memoization. The note
claimed that this was to cut recursion short, or something. I can't see
how it could recurse, given that `methods' is changing each time. Also,
the pair trick doesn't work with byte-compiled methods. So, remove it.
(compile-method): Dispatch to the appropriate method compiler, based on
whether the method was defined with the interpreter or with the
compiler.
(make-next-method): New function, generically computes a `next-method'
procedure, though the caller has to supply the arguments.
(compile-method/vm): Exciting method byte compiler!
(make-make-next-method/memoizer, compile-method/memoizer): Add the
/memoizer suffix, and move all this code to the bottom of the file.
* libguile/eval.h:
* libguile/eval.c (scm_make_promise): Rename from `scm_makprom', and
export as the scheme procedure, `make-promise'.
* libguile/eval.i.c (CEVAL): s/makprom/make_promise/.
