Filter refactoring: dropped the recursion from the interpreter

This is a major change of how the filters are interpreted. If everything
works how it should, it should not affect you unless you are hacking the
filters themselves.

Anyway, this change should make a huge improvement in the filter performance
as previous benchmarks showed that our major problem lies in the
recursion itself.

There are also some changes in nest and protocols, related mostly to
spreading const declarations throughout the whole BIRD and also to
refactored dynamic attribute definitions. The need of these came up
during the whole work and it is too difficult to split out these
not-so-related changes.

conf/confbase.Y

@@ -55,7 +55,6 @@ CF_DECLS
   struct f_tree *e;
   struct f_trie *trie;
   struct f_val v;
-  struct f_path_mask *h;
   struct password_item *p;
   struct rt_show_data *ra;
   struct sym_show_data *sd;
@@ -124,14 +123,14 @@ conf: definition ;
 definition:
    DEFINE SYM '=' term ';' {
      struct f_val *val = cfg_alloc(sizeof(struct f_val));
-     if (f_eval($4, cfg_mem, val) > F_RETURN) cf_error("Runtime error");
+     if (f_eval(f_postfixify($4), cfg_mem, val) > F_RETURN) cf_error("Runtime error");
      cf_define_symbol($2, SYM_CONSTANT | val->type, val);
    }
  ;
 
 expr:
    NUM
- | '(' term ')' { $$ = f_eval_int($2); }
+ | '(' term ')' { $$ = f_eval_int(f_postfixify($2)); }
  | SYM {
      if ($1->class != (SYM_CONSTANT | T_INT)) cf_error("Number expected");
      $$ = SYM_VAL($1).i; }

doc/bird.sgml

@@ -3982,7 +3982,7 @@ definitions, prefix definitions and DNS definitions:
 	RAdv protocol could be configured to change its behavior based on
 	availability of routes. When this option is used, the protocol waits in
 	suppressed state until a <it/trigger route/ (for the specified network)
-	is exported to the protocol, the protocol also returnsd to suppressed
+	is exported to the protocol, the protocol also returns to suppressed
 	state if the <it/trigger route/ disappears. Note that route export
 	depends on specified export filter, as usual. This option could be used,
 	e.g., for handling failover in multihoming scenarios.

filter/Makefile

@@ -5,10 +5,22 @@ $(cf-local)
 
 M4FLAGS_FILTERS=$(filter-out -s,$(M4FLAGS))
 
-$(o)f-inst-interpret.c: $(s)interpret.m4 $(s)f-inst.c
+$(o)f-inst-line-size.c: $(s)line-size.m4 $(s)f-inst.c $(objdir)/.dir-stamp
 	$(M4) $(M4FLAGS_FILTERS) -P $^ >$@
 
-$(o)filter.o: $(o)f-inst-interpret.c
+$(o)f-inst-postfixify.c: $(s)postfixify.m4 $(s)f-inst.c $(objdir)/.dir-stamp
+	$(M4) $(M4FLAGS_FILTERS) -P $^ >$@
+
+$(o)f-inst-interpret.c: $(s)interpret.m4 $(s)f-inst.c $(objdir)/.dir-stamp
+	$(M4) $(M4FLAGS_FILTERS) -P $^ >$@
+
+$(o)f-inst-same.c: $(s)same.m4 $(s)f-inst.c $(objdir)/.dir-stamp
+	$(M4) $(M4FLAGS_FILTERS) -P $^ >$@
+
+$(o)f-inst-dump.c: $(s)dump.m4 $(s)f-inst.c $(objdir)/.dir-stamp
+	$(M4) $(M4FLAGS_FILTERS) -P $^ >$@
+
+$(o)filter.o: $(o)f-inst-interpret.c $(o)f-inst-line-size.c $(o)f-inst-postfixify.c $(o)f-inst-same.c $(o)f-inst-dump.c
 
 tests_src := tree_test.c filter_test.c trie_test.c
 tests_targets := $(tests_targets) $(tests-target-files)