[PATCH] kconfig: use gperf for kconfig keywords

Use gperf to generate a hash for the kconfig keywords.  This greatly reduces
the size of the generated scanner and makes it easier to extend kconfig.

Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

scripts/kconfig/Makefile

@@ -114,7 +114,7 @@ gconf-objs	:= gconf.o kconfig_load.o zconf.tab.o
 endif
 
 clean-files	:= lkc_defs.h qconf.moc .tmp_qtcheck \
-		   .tmp_gtkcheck zconf.tab.c lex.zconf.c
+		   .tmp_gtkcheck zconf.tab.c lex.zconf.c zconf.hash.c
 
 # Needed for systems without gettext
 KBUILD_HAVE_NLS := $(shell \
@@ -201,7 +201,7 @@ $(obj)/.tmp_gtkcheck:
 	fi
 endif
 
-$(obj)/zconf.tab.o: $(obj)/lex.zconf.c
+$(obj)/zconf.tab.o: $(obj)/lex.zconf.c $(obj)/zconf.hash.c
 
 $(obj)/kconfig_load.o: $(obj)/lkc_defs.h
 
@@ -217,7 +217,7 @@ $(obj)/lkc_defs.h: $(src)/lkc_proto.h
 
 
 ###
-# The following requires flex/bison
+# The following requires flex/bison/gperf
 # By default we use the _shipped versions, uncomment the following line if
 # you are modifying the flex/bison src.
 # LKC_GENPARSER := 1
@@ -226,6 +226,7 @@ ifdef LKC_GENPARSER
 
 $(obj)/zconf.tab.c: $(src)/zconf.y
 $(obj)/lex.zconf.c: $(src)/zconf.l
+$(obj)/zconf.hash.c: $(src)/zconf.gperf
 
 %.tab.c: %.y
 	bison -l -b $* -p $(notdir $*) $<
@@ -235,4 +236,8 @@ lex.%.c: %.l
 	flex -L -P$(notdir $*) -o$@ $<
 	cp $@ $@_shipped
 
+%.hash.c: %.gperf
+	gperf < $< > $@
+	cp $@ $@_shipped
+
 endif

scripts/kconfig/lkc.h

@@ -37,6 +37,16 @@ extern "C" {
 #define _(text) gettext(text)
 #define N_(text) (text)
 
+
+#define TF_COMMAND	0x0001
+#define TF_PARAM	0x0002
+
+struct kconf_id {
+	int name;
+	int token;
+	unsigned int flags;
+};
+
 int zconfparse(void);
 void zconfdump(FILE *out);
 

scripts/kconfig/zconf.gperf

@@ -0,0 +1,43 @@
+%language=ANSI-C
+%define hash-function-name kconf_id_hash
+%define lookup-function-name kconf_id_lookup
+%define string-pool-name kconf_id_strings
+%compare-strncmp
+%enum
+%pic
+%struct-type
+
+struct kconf_id;
+
+%%
+mainmenu,	T_MAINMENU,	TF_COMMAND
+menu,		T_MENU,		TF_COMMAND
+endmenu,	T_ENDMENU,	TF_COMMAND
+source,		T_SOURCE,	TF_COMMAND
+choice,		T_CHOICE,	TF_COMMAND
+endchoice,	T_ENDCHOICE,	TF_COMMAND
+comment,	T_COMMENT,	TF_COMMAND
+config,		T_CONFIG,	TF_COMMAND
+menuconfig,	T_MENUCONFIG,	TF_COMMAND
+help,		T_HELP,		TF_COMMAND
+if,		T_IF,		TF_COMMAND|TF_PARAM
+endif,		T_ENDIF,	TF_COMMAND
+depends,	T_DEPENDS,	TF_COMMAND
+requires,	T_REQUIRES,	TF_COMMAND
+optional,	T_OPTIONAL,	TF_COMMAND
+default,	T_DEFAULT,	TF_COMMAND
+prompt,		T_PROMPT,	TF_COMMAND
+tristate,	T_TRISTATE,	TF_COMMAND
+def_tristate,	T_DEF_TRISTATE,	TF_COMMAND
+bool,		T_BOOLEAN,	TF_COMMAND
+boolean,	T_BOOLEAN,	TF_COMMAND
+def_bool,	T_DEF_BOOLEAN,	TF_COMMAND
+def_boolean,	T_DEF_BOOLEAN,	TF_COMMAND
+int,		T_INT,		TF_COMMAND
+hex,		T_HEX,		TF_COMMAND
+string,		T_STRING,	TF_COMMAND
+select,		T_SELECT,	TF_COMMAND
+enable,		T_SELECT,	TF_COMMAND
+range,		T_RANGE,	TF_COMMAND
+on,		T_ON,		TF_PARAM
+%%

scripts/kconfig/zconf.hash.c_shipped

@@ -0,0 +1,231 @@
+/* ANSI-C code produced by gperf version 3.0.1 */
+/* Command-line: gperf  */
+/* Computed positions: -k'1,3' */
+
+#if !((' ' == 32) && ('!' == 33) && ('"' == 34) && ('#' == 35) \
+      && ('%' == 37) && ('&' == 38) && ('\'' == 39) && ('(' == 40) \
+      && (')' == 41) && ('*' == 42) && ('+' == 43) && (',' == 44) \
+      && ('-' == 45) && ('.' == 46) && ('/' == 47) && ('0' == 48) \
+      && ('1' == 49) && ('2' == 50) && ('3' == 51) && ('4' == 52) \
+      && ('5' == 53) && ('6' == 54) && ('7' == 55) && ('8' == 56) \
+      && ('9' == 57) && (':' == 58) && (';' == 59) && ('<' == 60) \
+      && ('=' == 61) && ('>' == 62) && ('?' == 63) && ('A' == 65) \
+      && ('B' == 66) && ('C' == 67) && ('D' == 68) && ('E' == 69) \
+      && ('F' == 70) && ('G' == 71) && ('H' == 72) && ('I' == 73) \
+      && ('J' == 74) && ('K' == 75) && ('L' == 76) && ('M' == 77) \
+      && ('N' == 78) && ('O' == 79) && ('P' == 80) && ('Q' == 81) \
+      && ('R' == 82) && ('S' == 83) && ('T' == 84) && ('U' == 85) \
+      && ('V' == 86) && ('W' == 87) && ('X' == 88) && ('Y' == 89) \
+      && ('Z' == 90) && ('[' == 91) && ('\\' == 92) && (']' == 93) \
+      && ('^' == 94) && ('_' == 95) && ('a' == 97) && ('b' == 98) \
+      && ('c' == 99) && ('d' == 100) && ('e' == 101) && ('f' == 102) \
+      && ('g' == 103) && ('h' == 104) && ('i' == 105) && ('j' == 106) \
+      && ('k' == 107) && ('l' == 108) && ('m' == 109) && ('n' == 110) \
+      && ('o' == 111) && ('p' == 112) && ('q' == 113) && ('r' == 114) \
+      && ('s' == 115) && ('t' == 116) && ('u' == 117) && ('v' == 118) \
+      && ('w' == 119) && ('x' == 120) && ('y' == 121) && ('z' == 122) \
+      && ('{' == 123) && ('|' == 124) && ('}' == 125) && ('~' == 126))
+/* The character set is not based on ISO-646.  */
+#error "gperf generated tables don't work with this execution character set. Please report a bug to <bug-gnu-gperf@gnu.org>."
+#endif
+
+struct kconf_id;
+/* maximum key range = 45, duplicates = 0 */
+
+#ifdef __GNUC__
+__inline
+#else
+#ifdef __cplusplus
+inline
+#endif
+#endif
+static unsigned int
+kconf_id_hash (register const char *str, register unsigned int len)
+{
+  static unsigned char asso_values[] =
+    {
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 25, 10, 15,
+       0,  0,  5, 47,  0,  0, 47, 47,  0, 10,
+       0, 20, 20, 20,  5,  0,  0, 20, 47, 47,
+      20, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47, 47, 47, 47, 47,
+      47, 47, 47, 47, 47, 47
+    };
+  register int hval = len;
+
+  switch (hval)
+    {
+      default:
+        hval += asso_values[(unsigned char)str[2]];
+      /*FALLTHROUGH*/
+      case 2:
+      case 1:
+        hval += asso_values[(unsigned char)str[0]];
+        break;
+    }
+  return hval;
+}
+
+struct kconf_id_strings_t
+  {
+    char kconf_id_strings_str2[sizeof("if")];
+    char kconf_id_strings_str3[sizeof("int")];
+    char kconf_id_strings_str4[sizeof("help")];
+    char kconf_id_strings_str5[sizeof("endif")];
+    char kconf_id_strings_str6[sizeof("select")];
+    char kconf_id_strings_str7[sizeof("endmenu")];
+    char kconf_id_strings_str8[sizeof("tristate")];
+    char kconf_id_strings_str9[sizeof("endchoice")];
+    char kconf_id_strings_str10[sizeof("range")];
+    char kconf_id_strings_str11[sizeof("string")];
+    char kconf_id_strings_str12[sizeof("default")];
+    char kconf_id_strings_str13[sizeof("def_bool")];
+    char kconf_id_strings_str14[sizeof("menu")];
+    char kconf_id_strings_str16[sizeof("def_boolean")];
+    char kconf_id_strings_str17[sizeof("def_tristate")];
+    char kconf_id_strings_str18[sizeof("mainmenu")];
+    char kconf_id_strings_str20[sizeof("menuconfig")];
+    char kconf_id_strings_str21[sizeof("config")];
+    char kconf_id_strings_str22[sizeof("on")];
+    char kconf_id_strings_str23[sizeof("hex")];
+    char kconf_id_strings_str26[sizeof("source")];
+    char kconf_id_strings_str27[sizeof("depends")];
+    char kconf_id_strings_str28[sizeof("optional")];
+    char kconf_id_strings_str31[sizeof("enable")];
+    char kconf_id_strings_str32[sizeof("comment")];
+    char kconf_id_strings_str33[sizeof("requires")];
+    char kconf_id_strings_str34[sizeof("bool")];
+    char kconf_id_strings_str37[sizeof("boolean")];
+    char kconf_id_strings_str41[sizeof("choice")];
+    char kconf_id_strings_str46[sizeof("prompt")];
+  };
+static struct kconf_id_strings_t kconf_id_strings_contents =
+  {
+    "if",
+    "int",
+    "help",
+    "endif",
+    "select",
+    "endmenu",
+    "tristate",
+    "endchoice",
+    "range",
+    "string",
+    "default",
+    "def_bool",
+    "menu",
+    "def_boolean",
+    "def_tristate",
+    "mainmenu",
+    "menuconfig",
+    "config",
+    "on",
+    "hex",
+    "source",
+    "depends",
+    "optional",
+    "enable",
+    "comment",
+    "requires",
+    "bool",
+    "boolean",
+    "choice",
+    "prompt"
+  };
+#define kconf_id_strings ((const char *) &kconf_id_strings_contents)
+#ifdef __GNUC__
+__inline
+#endif
+struct kconf_id *
+kconf_id_lookup (register const char *str, register unsigned int len)
+{
+  enum
+    {
+      TOTAL_KEYWORDS = 30,
+      MIN_WORD_LENGTH = 2,
+      MAX_WORD_LENGTH = 12,
+      MIN_HASH_VALUE = 2,
+      MAX_HASH_VALUE = 46
+    };
+
+  static struct kconf_id wordlist[] =
+    {
+      {-1}, {-1},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str2,		T_IF,		TF_COMMAND|TF_PARAM},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str3,		T_INT,		TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str4,		T_HELP,		TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str5,		T_ENDIF,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str6,		T_SELECT,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str7,	T_ENDMENU,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str8,	T_TRISTATE,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str9,	T_ENDCHOICE,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str10,		T_RANGE,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str11,		T_STRING,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str12,	T_DEFAULT,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str13,	T_DEF_BOOLEAN,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str14,		T_MENU,		TF_COMMAND},
+      {-1},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str16,	T_DEF_BOOLEAN,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str17,	T_DEF_TRISTATE,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str18,	T_MAINMENU,	TF_COMMAND},
+      {-1},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str20,	T_MENUCONFIG,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str21,		T_CONFIG,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str22,		T_ON,		TF_PARAM},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str23,		T_HEX,		TF_COMMAND},
+      {-1}, {-1},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str26,		T_SOURCE,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str27,	T_DEPENDS,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str28,	T_OPTIONAL,	TF_COMMAND},
+      {-1}, {-1},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str31,		T_SELECT,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str32,	T_COMMENT,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str33,	T_REQUIRES,	TF_COMMAND},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str34,		T_BOOLEAN,	TF_COMMAND},
+      {-1}, {-1},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str37,	T_BOOLEAN,	TF_COMMAND},
+      {-1}, {-1}, {-1},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str41,		T_CHOICE,	TF_COMMAND},
+      {-1}, {-1}, {-1}, {-1},
+      {(int)(long)&((struct kconf_id_strings_t *)0)->kconf_id_strings_str46,		T_PROMPT,	TF_COMMAND}
+    };
+
+  if (len <= MAX_WORD_LENGTH && len >= MIN_WORD_LENGTH)
+    {
+      register int key = kconf_id_hash (str, len);
+
+      if (key <= MAX_HASH_VALUE && key >= 0)
+        {
+          register int o = wordlist[key].name;
+          if (o >= 0)
+            {
+              register const char *s = o + kconf_id_strings;
+
+              if (*str == *s && !strncmp (str + 1, s + 1, len - 1) && s[len] == '\0')
+                return &wordlist[key];
+            }
+        }
+    }
+  return 0;
+}
+

scripts/kconfig/zconf.l

@@ -18,8 +18,7 @@
 
 #define START_STRSIZE	16
 
-char *text;
-static char *text_ptr;
+static char *text;
 static int text_size, text_asize;
 
 struct buffer {
@@ -38,23 +37,22 @@ void new_string(void)
 {
 	text = malloc(START_STRSIZE);
 	text_asize = START_STRSIZE;
-	text_ptr = text;
 	text_size = 0;
-	*text_ptr = 0;
+	*text = 0;
 }
 
 void append_string(const char *str, int size)
 {
 	int new_size = text_size + size + 1;
 	if (new_size > text_asize) {
+		new_size += START_STRSIZE - 1;
+		new_size &= -START_STRSIZE;
 		text = realloc(text, new_size);
 		text_asize = new_size;
-		text_ptr = text + text_size;
 	}
-	memcpy(text_ptr, str, size);
-	text_ptr += size;
+	memcpy(text + text_size, str, size);
 	text_size += size;
-	*text_ptr = 0;
+	text[text_size] = 0;
 }
 
 void alloc_string(const char *str, int size)
@@ -88,36 +86,12 @@ n	[A-Za-z0-9_]
 
 
 <COMMAND>{
-	"mainmenu"		BEGIN(PARAM); return T_MAINMENU;
-	"menu"			BEGIN(PARAM); return T_MENU;
-	"endmenu"		BEGIN(PARAM); return T_ENDMENU;
-	"source"		BEGIN(PARAM); return T_SOURCE;
-	"choice"		BEGIN(PARAM); return T_CHOICE;
-	"endchoice"		BEGIN(PARAM); return T_ENDCHOICE;
-	"comment"		BEGIN(PARAM); return T_COMMENT;
-	"config"		BEGIN(PARAM); return T_CONFIG;
-	"menuconfig"		BEGIN(PARAM); return T_MENUCONFIG;
-	"help"			BEGIN(PARAM); return T_HELP;
-	"if"			BEGIN(PARAM); return T_IF;
-	"endif"			BEGIN(PARAM); return T_ENDIF;
-	"depends"		BEGIN(PARAM); return T_DEPENDS;
-	"requires"		BEGIN(PARAM); return T_REQUIRES;
-	"optional"		BEGIN(PARAM); return T_OPTIONAL;
-	"default"		BEGIN(PARAM); return T_DEFAULT;
-	"prompt"		BEGIN(PARAM); return T_PROMPT;
-	"tristate"		BEGIN(PARAM); return T_TRISTATE;
-	"def_tristate"		BEGIN(PARAM); return T_DEF_TRISTATE;
-	"bool"			BEGIN(PARAM); return T_BOOLEAN;
-	"boolean"		BEGIN(PARAM); return T_BOOLEAN;
-	"def_bool"		BEGIN(PARAM); return T_DEF_BOOLEAN;
-	"def_boolean"		BEGIN(PARAM); return T_DEF_BOOLEAN;
-	"int"			BEGIN(PARAM); return T_INT;
-	"hex"			BEGIN(PARAM); return T_HEX;
-	"string"		BEGIN(PARAM); return T_STRING;
-	"select"		BEGIN(PARAM); return T_SELECT;
-	"enable"		BEGIN(PARAM); return T_SELECT;
-	"range"			BEGIN(PARAM); return T_RANGE;
 	{n}+	{
+		struct kconf_id *id = kconf_id_lookup(yytext, yyleng);
+		if (id && id->flags & TF_COMMAND) {
+			BEGIN(PARAM);
+			return id->token;
+		}
 		alloc_string(yytext, yyleng);
 		zconflval.string = text;
 		return T_WORD;
@@ -134,8 +108,6 @@ n	[A-Za-z0-9_]
 	"!"	return T_NOT;
 	"="	return T_EQUAL;
 	"!="	return T_UNEQUAL;
-	"if"	return T_IF;
-	"on"	return T_ON;
 	\"|\'	{
 		str = yytext[0];
 		new_string();
@@ -144,6 +116,9 @@ n	[A-Za-z0-9_]
 	\n	BEGIN(INITIAL); current_file->lineno++; return T_EOL;
 	---	/* ignore */
 	({n}|[-/.])+	{
+		struct kconf_id *id = kconf_id_lookup(yytext, yyleng);
+		if (id && id->flags & TF_PARAM)
+			return id->token;
 		alloc_string(yytext, yyleng);
 		zconflval.string = text;
 		return T_WORD;

scripts/kconfig/zconf.y

@@ -11,6 +11,11 @@
 #include <string.h>
 #include <stdbool.h>
 
+#define LKC_DIRECT_LINK
+#include "lkc.h"
+
+#include "zconf.hash.c"
+
 #define printd(mask, fmt...) if (cdebug & (mask)) printf(fmt)
 
 #define PRINTD		0x0001
@@ -88,10 +93,6 @@ static struct menu *current_menu, *current_entry;
 %type <expr> if_expr
 %type <token> end
 
-%{
-#define LKC_DIRECT_LINK
-#include "lkc.h"
-%}
 %%
 input:	  /* empty */
 	| input block