* cppfiles.c (stack_include_file): Don't optimize zero-length

	files.
	(read_include_file): NUL-terminate read files.
	* cpplex.c (handle_newline, skip_escaped_newlines,
	get_effective_char, skip_whitespace, parse_identifier,
	parse_identifier_slow, parse_number, parse_string,
	_cpp_lex_direct): Optimize for the fact that buffers are guaranteed
	NUL-terminated.
	* cpplib.c (destringize_and_run, cpp_define, handle_assertion):
	Be sure buffers are NUL terminated.
	* cppmacro.c (warn_of_redefinition): Kill compile warning.

	* c-common.c: Include tree-inline.h.
	(c_language): Move separate definitions here.
	(c_common_init_options, c_common_post_options): New.
	(c_common_lang_init): Rename c_common_init.
	* c-common.h (c_common_lang_init): Similarly.
	(c_common_init_options, c_common_post_options): New.
	* c-lang.c (c_post_options): Move body to c_common_post_options.
	(c_init_options): Use c_common_init_options.
	(c_init): Update.
	* langhooks.def: Rearrange.
	* langhooks.h: Rearrange, and improve comments.
	* toplev.c (do_compile): New function.
	(toplev_main): Use it.
	(lang_independent_f_options, parse_options_and_default_flags,
	process_options): Remove trailing periods.
	* Makefile.in: Update.
cp:	* decl2.c (c_language): Move to c-common.c.
	* lex.c (cxx_post_options, cxx_init_options): Use c-common.c
	functions.
	(cxx_init): Update.
objc:	* objc-act.c (objc_post_options, objc_init_options): Use c-common.c
	functions.
	(ojbc_init): Update.


git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@47362 138bc75d-0d04-0410-961f-82ee72b054a4

16 files changed, 241 insertions, 178 deletions

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 1db9694fc80..496fd5aa723 100644
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,39 @@
+2001-11-26  Neil Booth  <neil@daikokuya.demon.co.uk>
+
+	* cppfiles.c (stack_include_file): Don't optimize zero-length
+	files.
+	(read_include_file): NUL-terminate read files.
+	* cpplex.c (handle_newline, skip_escaped_newlines,
+	get_effective_char, skip_whitespace, parse_identifier,
+	parse_identifier_slow, parse_number, parse_string,
+	_cpp_lex_direct): Optimize for the fact that buffers are guaranteed
+	NUL-terminated.
+	* cpplib.c (destringize_and_run, cpp_define, handle_assertion):
+	Be sure buffers are NUL terminated.
+	* cppmacro.c (warn_of_redefinition): Kill compile warning.
+
+2001-11-26  Neil Booth  <neil@daikokuya.demon.co.uk>
+
+	* c-common.c: Include tree-inline.h.
+	(c_language): Move separate definitions here.
+	(c_common_init_options, c_common_post_options): New.
+	(c_common_lang_init): Rename c_common_init.
+	* c-common.h (c_common_lang_init): Similarly.
+	(c_common_init_options, c_common_post_options): New.
+	* c-lang.c (c_post_options): Move body to c_common_post_options.
+	(c_init_options): Use c_common_init_options.
+	(c_init): Update.
+	* langhooks.def: Rearrange.
+	* langhooks.h: Rearrange, and improve comments.
+	* toplev.c (do_compile): New function.
+	(toplev_main): Use it.
+	(lang_independent_f_options, parse_options_and_default_flags,
+	process_options): Remove trailing periods.
+	* Makefile.in: Update.
+objc:	* objc-act.c (objc_post_options, objc_init_options): Use c-common.c
+	functions.
+	(ojbc_init): Update.
+
 2001-11-26  Richard Henderson  <rth@redhat.com>
 
 	* config/alpha/alpha.md (unop): Add 0 offset for some gas versions.
diff --git a/gcc/Makefile.in b/gcc/Makefile.in
index f88b0b4bf1e..1cb15998f73 100644
--- a/gcc/Makefile.in
+++ b/gcc/Makefile.in
@@ -1250,7 +1250,7 @@ s-under: $(GCC_PASSES)
 c-common.o : c-common.c $(CONFIG_H) $(SYSTEM_H) $(TREE_H) $(OBSTACK_H) \
 	$(C_COMMON_H) flags.h toplev.h output.h c-pragma.h $(RTL_H) $(GGC_H) \
 	$(EXPR_H) $(TM_P_H) builtin-types.def builtin-attrs.def $(TARGET_H) \
-	diagnostic.h
+	diagnostic.h tree-inline.h
 
 # A file used by all variants of C and some other languages.
 
diff --git a/gcc/c-common.c b/gcc/c-common.c
index dde1af3b892..a7d005b04ed 100644
--- a/gcc/c-common.c
+++ b/gcc/c-common.c
@@ -30,6 +30,7 @@ Software Foundation, 59 Temple Place - Suite 330, Boston, MA
 #include "ggc.h"
 #include "expr.h"
 #include "c-common.h"
+#include "tree-inline.h"
 #include "diagnostic.h"
 #include "tm_p.h"
 #include "obstack.h"
@@ -79,6 +80,10 @@ cpp_reader *parse_in;		/* Declared in c-lex.h.  */
 			: "long long unsigned int"))
 #endif
 
+/* The variant of the C language being processed.  */
+
+enum c_language_kind c_language;
+
 /* The following symbols are subsumed in the c_global_trees array, and
    listed here individually for documentation purposes.
 
@@ -2371,7 +2376,7 @@ c_common_nodes_and_builtins ()
   tree va_list_arg_type_node;
 
   /* We must initialize this before any builtin functions (which might have
-     attributes) are declared.  (c_common_lang_init is too late.)  */
+     attributes) are declared.  (c_common_init is too late.)  */
   format_attribute_table = c_format_attribute_table;
 
   /* Define `int' and `char' first so that dbx will output them first.  */
@@ -3858,17 +3863,43 @@ static bool c_attrs_initialized = false;
 
 static void c_init_attributes PARAMS ((void));
 
-/* Do the parts of lang_init common to C and C++.  */
-const char *
-c_common_lang_init (filename)
-     const char *filename;
+/* Common initialization before parsing options.  */
+void
+c_common_init_options (lang)
+     enum c_language_kind lang;
 {
-  filename = init_c_lex (filename);
+  c_language = lang;
+  parse_in = cpp_create_reader (lang == clk_c ? CLK_GNUC89:
+				lang == clk_cplusplus ? CLK_GNUCXX: CLK_OBJC);
 
-  init_pragma ();
+  /* Mark as "unspecified" (see c_common_post_options).  */
+  flag_bounds_check = -1;
+}
+
+/* Post-switch processing.  */
+void
+c_common_post_options ()
+{
+  cpp_post_options (parse_in);
+
+  /* Use tree inlining if possible.  Function instrumentation is only
+     done in the RTL level, so we disable tree inlining.  */
+  if (! flag_instrument_function_entry_exit)
+    {
+      if (!flag_no_inline)
+	{
+	  flag_inline_trees = 1;
+	  flag_no_inline = 1;
+	}
+      if (flag_inline_functions)
+	{
+	  flag_inline_trees = 2;
+	  flag_inline_functions = 0;
+	}
+    }
 
   /* If still "unspecified", make it match -fbounded-pointers.  */
-  if (flag_bounds_check < 0)
+  if (flag_bounds_check == -1)
     flag_bounds_check = flag_bounded_pointers;
 
   /* Special format checking options don't work without -Wformat; warn if
@@ -3883,6 +3914,18 @@ c_common_lang_init (filename)
     warning ("-Wformat-security ignored without -Wformat");
   if (warn_missing_format_attribute && !warn_format)
     warning ("-Wmissing-format-attribute ignored without -Wformat");
+}
+
+/* Front end initialization common to C, ObjC and C++.  */
+const char *
+c_common_init (filename)
+     const char *filename;
+{
+  /* Do this before initializing pragmas, as then cpplib's hash table
+     has been set up.  */
+  filename = init_c_lex (filename);
+
+  init_pragma ();
 
   if (!c_attrs_initialized)
     c_init_attributes ();
diff --git a/gcc/c-common.h b/gcc/c-common.h
index b289cb8b24f..84e4bb95824 100644
--- a/gcc/c-common.h
+++ b/gcc/c-common.h
@@ -534,7 +534,9 @@ extern void disable_builtin_function		PARAMS ((const char *));
 
 extern tree build_va_arg			PARAMS ((tree, tree));
 
-extern const char *c_common_lang_init		PARAMS ((const char *));
+extern void c_common_init_options		PARAMS ((enum c_language_kind));
+extern void c_common_post_options		PARAMS ((void));
+extern const char *c_common_init		PARAMS ((const char *));
 extern void c_common_finish			PARAMS ((void));
 extern HOST_WIDE_INT c_common_get_alias_set	PARAMS ((tree));
 extern bool c_promoting_integer_type_p		PARAMS ((tree));
diff --git a/gcc/c-lang.c b/gcc/c-lang.c
index 9f0fb286a88..a5c86625b30 100644
--- a/gcc/c-lang.c
+++ b/gcc/c-lang.c
@@ -88,32 +88,13 @@ static varray_type deferred_fns;
 static void
 c_post_options ()
 {
-  cpp_post_options (parse_in);
-
-  /* Use tree inlining if possible.  Function instrumentation is only
-     done in the RTL level, so we disable tree inlining.  */
-  if (! flag_instrument_function_entry_exit)
-    {
-      if (!flag_no_inline)
-	{
-	  flag_inline_trees = 1;
-	  flag_no_inline = 1;
-	}
-      if (flag_inline_functions)
-	{
-	  flag_inline_trees = 2;
-	  flag_inline_functions = 0;
-	}
-    }
+  c_common_post_options ();
 }
 
 static void
 c_init_options ()
 {
-  parse_in = cpp_create_reader (CLK_GNUC89);
-
-  /* Mark as "unspecified".  */
-  flag_bounds_check = -1;
+  c_common_init_options (clk_c);
 }
 
 static const char *
@@ -122,7 +103,7 @@ c_init (filename)
 {
   c_init_decl_processing ();
 
-  filename = c_common_lang_init (filename);
+  filename = c_common_init (filename);
 
   add_c_tree_codes ();
 
diff --git a/gcc/cp/ChangeLog b/gcc/cp/ChangeLog
index 68ae9e1cfc6..110f7c40e8e 100644
--- a/gcc/cp/ChangeLog
+++ b/gcc/cp/ChangeLog
@@ -1,3 +1,10 @@
+2001-11-26  Neil Booth  <neil@daikokuya.demon.co.uk>
+
+	* decl2.c (c_language): Move to c-common.c.
+	* lex.c (cxx_post_options, cxx_init_options): Use c-common.c
+	functions.
+	(cxx_init): Update.
+
 2001-11-26  Jason Merrill  <jason@redhat.com>
 
 	* call.c (joust): Remove COND_EXPR hack.
diff --git a/gcc/cp/decl2.c b/gcc/cp/decl2.c
index f2bbba09f2c..c316da034fa 100644
--- a/gcc/cp/decl2.c
+++ b/gcc/cp/decl2.c
@@ -400,10 +400,6 @@ int flag_permissive;
 
 int flag_enforce_eh_specs = 1;
 
-/* The variant of the C language being processed.  */
-
-c_language_kind c_language = clk_cplusplus;
-
 /* Table of language-dependent -f options.
    STRING is the option name.  VARIABLE is the address of the variable.
    ON_VALUE is the value to store in VARIABLE
diff --git a/gcc/cp/lex.c b/gcc/cp/lex.c
index c093fc1f0f7..ba93ca5f5fa 100644
--- a/gcc/cp/lex.c
+++ b/gcc/cp/lex.c
@@ -241,18 +241,17 @@ static const char *const cplus_tree_code_name[] = {
 void
 cxx_post_options ()
 {
-  cpp_post_options (parse_in);
+  c_common_post_options ();
 }
 
+/* Initialization before switch parsing.  */
 void
 cxx_init_options ()
 {
-  parse_in = cpp_create_reader (CLK_GNUCXX);
+  c_common_init_options (clk_cplusplus);
 
   /* Default exceptions on.  */
   flag_exceptions = 1;
-  /* Mark as "unspecified".  */
-  flag_bounds_check = -1;
   /* By default wrap lines at 80 characters.  Is getenv ("COLUMNS")
      preferable?  */
   diagnostic_line_cutoff (global_dc) = 80;
@@ -720,10 +719,7 @@ cxx_init (filename)
 
   cxx_init_decl_processing ();
 
-  /* Create the built-in __null node.  Note that we can't yet call for
-     type_for_size here because integer_type_node and so forth are not
-     set up.  Therefore, we don't set the type of these nodes until
-     cxx_init_decl_processing.  */
+  /* Create the built-in __null node.  */
   null_node = build_int_2 (0, 0);
   TREE_TYPE (null_node) = type_for_size (POINTER_SIZE, 0);
   ridpointers[RID_NULL] = null_node;
@@ -731,7 +727,7 @@ cxx_init (filename)
   token_count = init_cpp_parse ();
   interface_unknown = 1;
 
-  filename = c_common_lang_init (filename);
+  filename = c_common_init (filename);
 
   init_cp_pragma ();
 
diff --git a/gcc/cppfiles.c b/gcc/cppfiles.c
index 1fb357d642a..f0d85d9bf37 100644
--- a/gcc/cppfiles.c
+++ b/gcc/cppfiles.c
@@ -296,17 +296,17 @@ stack_include_file (pfile, inc)
   /* Not in cache?  */
   if (! inc->buffer)
     {
-      /* Mark a regular, zero-length file never-reread.  Zero-length
-	 files are stacked the first time, so preprocessing a main
-	 file of zero length does not raise an error.  */
-      if (S_ISREG (inc->st.st_mode) && inc->st.st_size == 0)
-	_cpp_never_reread (inc);
-      else if (read_include_file (pfile, inc))
+      if (read_include_file (pfile, inc))
 	{
 	  /* If an error occurs, do not try to read this file again.  */
 	  _cpp_never_reread (inc);
 	  return false;
 	}
+      /* Mark a regular, zero-length file never-reread.  We read it,
+	 NUL-terminate it, and stack it once, so preprocessing a main
+	 file of zero length does not raise an error.  */
+      if (S_ISREG (inc->st.st_mode) && inc->st.st_size == 0)
+	_cpp_never_reread (inc);
       close (inc->fd);
       inc->fd = -1;
     }
@@ -382,7 +382,8 @@ read_include_file (pfile, inc)
       if (pagesize == -1)
 	pagesize = getpagesize ();
 
-      if (size / pagesize >= MMAP_THRESHOLD)
+      if (size / pagesize >= MMAP_THRESHOLD
+	  && (size % pagesize) != 0)
 	{
 	  buf = (U_CHAR *) mmap (0, size, PROT_READ, MAP_PRIVATE, inc->fd, 0);
 	  if (buf == (U_CHAR *)-1)
@@ -392,7 +393,7 @@ read_include_file (pfile, inc)
       else
 #endif
 	{
-	  buf = (U_CHAR *) xmalloc (size);
+	  buf = (U_CHAR *) xmalloc (size + 1);
 	  offset = 0;
 	  while (offset < size)
 	    {
@@ -410,6 +411,8 @@ read_include_file (pfile, inc)
 		}
 	      offset += count;
 	    }
+	  /* The lexer requires that the buffer be NUL-terminated.  */
+	  buf[size] = '\0';
 	}
     }
   else if (S_ISBLK (inc->st.st_mode))
@@ -424,19 +427,25 @@ read_include_file (pfile, inc)
 	 bigger than the majority of C source files.  */
       size = 8 * 1024;
 
-      buf = (U_CHAR *) xmalloc (size);
+      buf = (U_CHAR *) xmalloc (size + 1);
       offset = 0;
       while ((count = read (inc->fd, buf + offset, size - offset)) > 0)
 	{
 	  offset += count;
 	  if (offset == size)
-	    buf = xrealloc (buf, (size *= 2));
+	    {
+	      size *= 2;
+	      buf = xrealloc (buf, size + 1);
+	    }
 	}
       if (count < 0)
 	goto perror_fail;
 
-      if (offset < size)
-	buf = xrealloc (buf, offset);
+      if (offset + 1 < size)
+	buf = xrealloc (buf, offset + 1);
+
+      /* The lexer requires that the buffer be NUL-terminated.  */
+      buf[offset] = '\0';
       inc->st.st_size = offset;
     }
 
diff --git a/gcc/cpplex.c b/gcc/cpplex.c
index 35411e615c1..08223bd49b7 100644
--- a/gcc/cpplex.c
+++ b/gcc/cpplex.c
@@ -75,7 +75,7 @@ static cppchar_t get_effective_char PARAMS ((cpp_reader *));
 static int skip_block_comment PARAMS ((cpp_reader *));
 static int skip_line_comment PARAMS ((cpp_reader *));
 static void adjust_column PARAMS ((cpp_reader *));
-static void skip_whitespace PARAMS ((cpp_reader *, cppchar_t));
+static int skip_whitespace PARAMS ((cpp_reader *, cppchar_t));
 static cpp_hashnode *parse_identifier PARAMS ((cpp_reader *));
 static cpp_hashnode *parse_identifier_slow PARAMS ((cpp_reader *,
 						    const U_CHAR *));
@@ -119,12 +119,8 @@ handle_newline (pfile)
   cpp_buffer *buffer = pfile->buffer;
 
   /* Handle CR-LF and LF-CR.  Most other implementations (e.g. java)
-     only accept CR-LF; maybe we should fall back to that behaviour?
-
-     NOTE: the EOF case in _cpp_lex_direct currently requires the
-     buffer->cur != buffer->rlimit test here for 0-length files.  */
-  if (buffer->cur != buffer->rlimit
-      && buffer->cur[-1] + buffer->cur[0] == '\r' + '\n')
+     only accept CR-LF; maybe we should fall back to that behaviour?  */
+  if (buffer->cur[-1] + buffer->cur[0] == '\r' + '\n')
     buffer->cur++;
 
   buffer->line_base = buffer->cur;
@@ -190,24 +186,21 @@ skip_escaped_newlines (pfile)
 
       do
 	{
-	  if (buffer->cur == buffer->rlimit)
-	    break;
-      
 	  if (next == '?')
 	    {
-	      if (buffer->cur[0] != '?' || buffer->cur + 1 == buffer->rlimit)
-		break;
-
-	      if (!trigraph_p (pfile))
+	      if (buffer->cur[0] != '?' || !trigraph_p (pfile))
 		break;
 
 	      /* Translate the trigraph.  */
 	      next = _cpp_trigraph_map[buffer->cur[1]];
 	      buffer->cur += 2;
-	      if (next != '\\' || buffer->cur == buffer->rlimit)
+	      if (next != '\\')
 		break;
 	    }
 
+	  if (buffer->cur == buffer->rlimit)
+	    break;
+
 	  /* We have a backslash, and room for at least one more
 	     character.  Skip horizontal whitespace.  */
 	  saved_cur = buffer->cur;
@@ -250,16 +243,13 @@ static cppchar_t
 get_effective_char (pfile)
      cpp_reader *pfile;
 {
-  cppchar_t next = EOF;
+  cppchar_t next;
   cpp_buffer *buffer = pfile->buffer;
 
   buffer->backup_to = buffer->cur;
-  if (buffer->cur < buffer->rlimit)
-    {
-      next = *buffer->cur++;
-      if (__builtin_expect (next == '?' || next == '\\', 0))
-	next = skip_escaped_newlines (pfile);
-    }
+  next = *buffer->cur++;
+  if (__builtin_expect (next == '?' || next == '\\', 0))
+    next = skip_escaped_newlines (pfile);
 
    return next;
 }
@@ -295,7 +285,6 @@ skip_block_comment (pfile)
 	     comes immediately before the true comment delimiter.
 	     Don't bother to get it right across escaped newlines.  */
 	  if (CPP_OPTION (pfile, warn_comments)
-	      && buffer->cur + 1 < buffer->rlimit
 	      && buffer->cur[0] == '*' && buffer->cur[1] != '/')
 	    cpp_warning_with_line (pfile,
 				   pfile->line, CPP_BUF_COL (buffer),
@@ -360,7 +349,7 @@ adjust_column (pfile)
 /* Skips whitespace, saving the next non-whitespace character.
    Adjusts pfile->col_adjust to account for tabs.  Without this,
    tokens might be assigned an incorrect column.  */
-static void
+static int
 skip_whitespace (pfile, c)
      cpp_reader *pfile;
      cppchar_t c;
@@ -378,6 +367,8 @@ skip_whitespace (pfile, c)
       /* Just \f \v or \0 left.  */
       else if (c == '\0')
 	{
+	  if (buffer->cur - 1 == buffer->rlimit)
+	    return 0;
 	  if (!warned)
 	    {
 	      cpp_warning (pfile, "null character(s) ignored");
@@ -390,14 +381,13 @@ skip_whitespace (pfile, c)
 			       "%s in preprocessing directive",
 			       c == '\f' ? "form feed" : "vertical tab");
 
-      if (buffer->cur == buffer->rlimit)
-	return;
       c = *buffer->cur++;
     }
   /* We only want non-vertical space, i.e. ' ' \t \f \v \0.  */
   while (is_nvspace (c));
 
   buffer->cur--;
+  return 1;
 }
 
 /* See if the characters of a number token are valid in a name (no
@@ -430,18 +420,16 @@ parse_identifier (pfile)
      cpp_reader *pfile;
 {
   cpp_hashnode *result;
-  const U_CHAR *cur, *rlimit;
+  const U_CHAR *cur;
 
   /* Fast-path loop.  Skim over a normal identifier.
      N.B. ISIDNUM does not include $.  */
-  cur    = pfile->buffer->cur - 1;
-  rlimit = pfile->buffer->rlimit;
-  do
+  cur = pfile->buffer->cur;
+  while (ISIDNUM (*cur))
     cur++;
-  while (cur < rlimit && ISIDNUM (*cur));
 
   /* Check for slow-path cases.  */
-  if (cur < rlimit && (*cur == '?' || *cur == '\\' || *cur == '$'))
+  if (*cur == '?' || *cur == '\\' || *cur == '$')
     result = parse_identifier_slow (pfile, cur);
   else
     {
@@ -501,9 +489,6 @@ parse_identifier_slow (pfile, cur)
           if (c == '$')
             saw_dollar++;
 
-          if (buffer->cur == buffer->rlimit)
-            goto at_eof;
-
           c = *buffer->cur++;
         }
 
@@ -515,9 +500,8 @@ parse_identifier_slow (pfile, cur)
     }
   while (is_idchar (c));
 
-  /* Step back over the unwanted char, except at EOF.  */
+  /* Step back over the unwanted char.  */
   BACKUP ();
- at_eof:
 
   /* $ is not an identifier character in the standard, but is commonly
      accepted as an extension.  Don't warn about it in skipped
@@ -573,9 +557,6 @@ parse_number (pfile, number, c, leading_period)
 	    }
 	  *dest++ = c;
 
-	  if (buffer->cur == buffer->rlimit)
-	    goto at_eof;
-
 	  c = *buffer->cur++;
 	}
       while (is_numchar (c) || c == '.' || VALID_SIGN (c, dest[-1]));
@@ -588,9 +569,8 @@ parse_number (pfile, number, c, leading_period)
     }
   while (is_numchar (c) || c == '.' || VALID_SIGN (c, dest[-1]));
 
-  /* Step back over the unwanted char, except at EOF.  */
+  /* Step back over the unwanted char.  */
   BACKUP ();
- at_eof:
 
   /* Null-terminate the number.  */
   *dest = '\0';
@@ -671,12 +651,6 @@ parse_string (pfile, token, terminator)
 	  limit = BUFF_LIMIT (pfile->u_buff);
 	}
 
-      if (buffer->cur == buffer->rlimit)
-	{
-	  unterminated (pfile, terminator);
-	  break;
-	}
-
       /* Handle trigraphs, escaped newlines etc.  */
       c = *buffer->cur++;
       if (c == '?' || c == '\\')
@@ -724,10 +698,19 @@ parse_string (pfile, token, terminator)
 	  handle_newline (pfile);
 	  c = '\n';
 	}
-      else if (c == '\0' && !warned_nulls)
+      else if (c == '\0')
 	{
-	  warned_nulls = true;
-	  cpp_warning (pfile, "null character(s) preserved in literal");
+	  if (buffer->cur - 1 == buffer->rlimit)
+	    {
+	      unterminated (pfile, terminator);
+	      buffer->cur--;
+	      break;
+	    }
+	  if (!warned_nulls)
+	    {
+	      warned_nulls = true;
+	      cpp_warning (pfile, "null character(s) preserved in literal");
+	    }
 	}
 
       *dest++ = c;
@@ -907,15 +890,19 @@ _cpp_lex_direct (pfile)
   result->line = pfile->line;
 
  skipped_white:
-  if (buffer->cur == buffer->rlimit)
-    goto at_eof;
   c = *buffer->cur++;
   result->col = CPP_BUF_COLUMN (buffer, buffer->cur);
 
  trigraph:
   switch (c)
     {
-    at_eof:
+    case ' ': case '\t': case '\f': case '\v': case '\0':
+      result->flags |= PREV_WHITE;
+      if (skip_whitespace (pfile, c))
+	goto skipped_white;
+
+      /* EOF.  */
+      buffer->cur--;
       buffer->saved_flags = BOL;
       if (!pfile->state.parsing_args && !pfile->state.in_directive)
 	{
@@ -941,11 +928,6 @@ _cpp_lex_direct (pfile)
       result->type = CPP_EOF;
       break;
 
-    case ' ': case '\t': case '\f': case '\v': case '\0':
-      skip_whitespace (pfile, c);
-      result->flags |= PREV_WHITE;
-      goto skipped_white;
-
     case '\n': case '\r':
       handle_newline (pfile);
       buffer->saved_flags = BOL;
@@ -1016,8 +998,7 @@ _cpp_lex_direct (pfile)
       result->val.node = parse_identifier (pfile);
 
       /* 'L' may introduce wide characters or strings.  */
-      if (result->val.node == pfile->spec_nodes.n_L
-	  && buffer->cur < buffer->rlimit)
+      if (result->val.node == pfile->spec_nodes.n_L)
 	{
 	  c = *buffer->cur;
 	  if (c == '\'' || c == '"')
diff --git a/gcc/cpplib.c b/gcc/cpplib.c
index 324c7b21c4d..ea1d9f86860 100644
--- a/gcc/cpplib.c
+++ b/gcc/cpplib.c
@@ -1158,7 +1158,7 @@ destringize_and_run (pfile, in)
   const unsigned char *src, *limit;
   char *dest, *result;
 
-  dest = result = alloca (in->len);
+  dest = result = alloca (in->len + 1);
   for (src = in->text, limit = src + in->len; src < limit;)
     {
       /* We know there is a character following the backslash.  */
@@ -1166,6 +1166,7 @@ destringize_and_run (pfile, in)
 	src++;
       *dest++ = *src++;
     }
+  *dest = '\0';
 
   run_directive (pfile, T_PRAGMA, result, dest - result);
 }
@@ -1647,9 +1648,8 @@ cpp_define (pfile, str)
      Change the first "=" in the string to a space.  If there is none,
      tack " 1" on the end.  */
 
-  /* Length including the null.  */  
   count = strlen (str);
-  buf = (char *) alloca (count + 2);
+  buf = (char *) alloca (count + 3);
   memcpy (buf, str, count);
 
   p = strchr (str, '=');
@@ -1660,6 +1660,7 @@ cpp_define (pfile, str)
       buf[count++] = ' ';
       buf[count++] = '1';
     }
+  buf[count] = '\0';
 
   run_directive (pfile, T_DEFINE, buf, count);
 }
@@ -1714,11 +1715,12 @@ handle_assertion (pfile, str, type)
     {
       /* Copy the entire option so we can modify it.  Change the first
 	 "=" in the string to a '(', and tack a ')' on the end.  */
-      char *buf = (char *) alloca (count + 1);
+      char *buf = (char *) alloca (count + 2);
 
       memcpy (buf, str, count);
       buf[p - str] = '(';
       buf[count++] = ')';
+      buf[count] = '\0';
       str = buf;
     }
 
diff --git a/gcc/cppmacro.c b/gcc/cppmacro.c
index 403920c03a3..21d3cb1296b 100644
--- a/gcc/cppmacro.c
+++ b/gcc/cppmacro.c
@@ -82,7 +82,7 @@ static _cpp_buff *funlike_invocation_p PARAMS ((cpp_reader *, cpp_hashnode *));
 
 static cpp_token *alloc_expansion_token PARAMS ((cpp_reader *, cpp_macro *));
 static cpp_token *lex_expansion_token PARAMS ((cpp_reader *, cpp_macro *));
-static int warn_of_redefinition PARAMS ((cpp_reader *, const cpp_hashnode *,
+static int warn_of_redefinition PARAMS ((const cpp_hashnode *,
 					 const cpp_macro *));
 static int save_parameter PARAMS ((cpp_reader *, cpp_macro *, cpp_hashnode *));
 static int parse_params PARAMS ((cpp_reader *, cpp_macro *));
@@ -380,6 +380,7 @@ paste_tokens (pfile, plhs, rhs)
       && (rhs->type == CPP_MULT || rhs->type == CPP_DIV))
     *end++ = ' ';
   end = cpp_spell_token (pfile, rhs, end);
+  *end = '\0';
 
   cpp_push_buffer (pfile, buf, end - buf, /* from_stage3 */ true, 1);
 
@@ -1112,8 +1113,7 @@ _cpp_backup_tokens (pfile, count)
 
 /* Returns non-zero if a macro redefinition warning is required.  */
 static int
-warn_of_redefinition (pfile, node, macro2)
-     cpp_reader *pfile;
+warn_of_redefinition (node, macro2)
      const cpp_hashnode *node;
      const cpp_macro *macro2;
 {
@@ -1408,7 +1408,7 @@ _cpp_create_definition (pfile, node)
 
   if (node->type != NT_VOID)
     {
-      if (warn_of_redefinition (pfile, node, macro))
+      if (warn_of_redefinition (node, macro))
 	{
 	  cpp_pedwarn_with_line (pfile, pfile->directive_line, 0,
 				 "\"%s\" redefined", NODE_NAME (node));
diff --git a/gcc/langhooks-def.h b/gcc/langhooks-def.h
index 7c8480f4f84..bb38e01cc71 100644
--- a/gcc/langhooks-def.h
+++ b/gcc/langhooks-def.h
@@ -119,12 +119,12 @@ int lhd_tree_dump_type_quals			PARAMS ((tree));
 #define LANG_HOOKS_INITIALIZER { \
   LANG_HOOKS_NAME, \
   LANG_HOOKS_IDENTIFIER_SIZE, \
-  LANG_HOOKS_INIT, \
-  LANG_HOOKS_FINISH, \
-  LANG_HOOKS_CLEAR_BINDING_STACK, \
   LANG_HOOKS_INIT_OPTIONS, \
   LANG_HOOKS_DECODE_OPTION, \
   LANG_HOOKS_POST_OPTIONS, \
+  LANG_HOOKS_INIT, \
+  LANG_HOOKS_FINISH, \
+  LANG_HOOKS_CLEAR_BINDING_STACK, \
   LANG_HOOKS_GET_ALIAS_SET, \
   LANG_HOOKS_HONOR_READONLY, \
   LANG_HOOKS_PRINT_STATISTICS, \
diff --git a/gcc/langhooks.h b/gcc/langhooks.h
index 4ec01e70d01..befdb0f6838 100644
--- a/gcc/langhooks.h
+++ b/gcc/langhooks.h
@@ -71,21 +71,8 @@ struct lang_hooks
      identifier nodes long enough for the language-specific slots.  */
   size_t identifier_size;
 
-  /* Called after options parsing, to initialize the front end.  The
-     main input filename is passed, which may be NULL; the front end
-     should return the original filename (e.g. foo.i -> foo.c).
-     Return NULL to indicate a serious error of some sort; in that
-     case no compilation is performed, and the finish hook is called
-     immediately.  */
-  const char * (*init) PARAMS ((const char *));
-
-  /* Called last, as a finalizer.  */
-  void (*finish) PARAMS ((void));
-
-  /* Called immediately after parsing to clear the binding stack.  */
-  void (*clear_binding_stack) PARAMS ((void));
-
-  /* Called to initialize options, before any calls to decode_option.  */
+  /* The first callback made to the front end, for simple
+     initialization needed before any calls to decode_option.  */
   void (*init_options) PARAMS ((void));
 
   /* Function called with an option vector as argument, to decode a
@@ -98,9 +85,26 @@ struct lang_hooks
      done for this option.  */
   int (*decode_option) PARAMS ((int, char **));
 
-  /* Called when all command line options have been parsed.  */
+  /* Called when all command line options have been parsed.  Should do
+     any required consistency checks, modifications etc.  Complex
+     initialization should be left to the "init" callback, since GC
+     and the identifier hashes are set up between now and then.  */
   void (*post_options) PARAMS ((void));
 
+  /* Called after post_options, to initialize the front end.  The main
+     input filename is passed, which may be NULL; the front end should
+     return the original filename (e.g. foo.i -> foo.c).  Return NULL
+     to indicate a serious error of some sort; in that case no
+     compilation is performed, and the finish hook is called
+     immediately.  */
+  const char * (*init) PARAMS ((const char *));
+
+  /* Called at the end of compilation, as a finalizer.  */
+  void (*finish) PARAMS ((void));
+
+  /* Called immediately after parsing to clear the binding stack.  */
+  void (*clear_binding_stack) PARAMS ((void));
+
   /* Called to obtain the alias set to be used for an expression or type.
      Returns -1 if the language does nothing special for it.  */
   HOST_WIDE_INT (*get_alias_set) PARAMS ((tree));
diff --git a/gcc/objc/objc-act.c b/gcc/objc/objc-act.c
index c98b61f02ab..474df1afdba 100644
--- a/gcc/objc/objc-act.c
+++ b/gcc/objc/objc-act.c
@@ -478,7 +478,7 @@ static varray_type deferred_fns;
 static void
 objc_post_options ()
 {
-  cpp_post_options (parse_in);
+  c_common_post_options ();
 }
 
 /* Some platforms pass small structures through registers versus through
@@ -547,8 +547,7 @@ generate_struct_by_value_array ()
 static void
 objc_init_options ()
 {
-  parse_in = cpp_create_reader (CLK_OBJC);
-  c_language = clk_objective_c;
+  c_common_init_options (clk_objective_c);
 }
 
 static const char *
@@ -557,7 +556,7 @@ objc_init (filename)
 {
   c_init_decl_processing ();
 
-  filename = c_common_lang_init (filename);
+  filename = c_common_init (filename);
 
   add_c_tree_codes ();
 
diff --git a/gcc/toplev.c b/gcc/toplev.c
index 7ee39cba43c..72c93a799b2 100644
--- a/gcc/toplev.c
+++ b/gcc/toplev.c
@@ -126,6 +126,7 @@ extern tree last_assemble_variable_decl;
 
 static void general_init PARAMS ((char *));
 static void parse_options_and_default_flags PARAMS ((int, char **));
+static void do_compile PARAMS ((void));
 static void process_options PARAMS ((void));
 static void lang_independent_init PARAMS ((void));
 static int lang_dependent_init PARAMS ((const char *));
@@ -1177,7 +1178,7 @@ lang_independent_options f_options[] =
   {"mem-report", &mem_report, 1,
    N_("Report on permanent memory allocation at end of run") },
   { "trapv", &flag_trapv, 1,
-   N_("Trap for signed overflow in addition / subtraction / multiplication.") },
+   N_("Trap for signed overflow in addition / subtraction / multiplication") },
 };
 
 /* Table of language-specific options.  */
@@ -4787,7 +4788,8 @@ parse_options_and_default_flags (argc, argv)
 	}
     }
 
-  /* All command line options have been processed.  */
+  /* All command line options have been parsed; allow the front end to
+     perform consistency checks, etc.  */
   (*lang_hooks.post_options) ();
 }
 
@@ -4837,7 +4839,7 @@ process_options ()
 
   if (profile_block_flag == 3)
     {
-      warning ("`-ax' and `-a' are conflicting options. `-a' ignored.");
+      warning ("`-ax' and `-a' are conflicting options. `-a' ignored");
       profile_block_flag = 2;
     }
 
@@ -4951,12 +4953,12 @@ process_options ()
     {
       if (flag_function_sections)
 	{
-	  warning ("-ffunction-sections not supported for this target.");
+	  warning ("-ffunction-sections not supported for this target");
 	  flag_function_sections = 0;
 	}
       if (flag_data_sections)
 	{
-	  warning ("-fdata-sections not supported for this target.");
+	  warning ("-fdata-sections not supported for this target");
 	  flag_data_sections = 0;
 	}
     }
@@ -4964,13 +4966,13 @@ process_options ()
   if (flag_function_sections
       && (profile_flag || profile_block_flag))
     {
-      warning ("-ffunction-sections disabled; it makes profiling impossible.");
+      warning ("-ffunction-sections disabled; it makes profiling impossible");
       flag_function_sections = 0;
     }
 
 #ifndef OBJECT_FORMAT_ELF
   if (flag_function_sections && write_symbols != NO_DEBUG)
-    warning ("-ffunction-sections may affect debugging on some targets.");
+    warning ("-ffunction-sections may affect debugging on some targets");
 #endif
 }
 
@@ -5127,29 +5129,10 @@ finalize ()
   (*lang_hooks.finish) ();
 }
 
-/* Entry point of cc1, cc1plus, jc1, f771, etc.
-   Decode command args, then call compile_file.
-   Exit code is FATAL_EXIT_CODE if can't open files or if there were
-   any errors, or SUCCESS_EXIT_CODE if compilation succeeded.
-
-   It is not safe to call this function more than once.  */
-
-int
-toplev_main (argc, argv)
-     int argc;
-     char **argv;
+/* Initialize the compiler, and compile the input file.  */
+static void
+do_compile ()
 {
-  /* Initialization of GCC's environment, and diagnostics.  */
-  general_init (argv [0]);
-
-  /* Parse the options and do minimal processing; basically just
-     enough to default flags appropriately.  */
-  parse_options_and_default_flags (argc, argv);
-
-  /* Exit early if we can (e.g. -help).  */
-  if (exit_after_options)
-    return (SUCCESS_EXIT_CODE);
-
   /* The bulk of command line switch processing.  */
   process_options ();
 
@@ -5171,6 +5154,30 @@ toplev_main (argc, argv)
   /* Stop timing and print the times.  */
   timevar_stop (TV_TOTAL);
   timevar_print (stderr);
+}
+
+/* Entry point of cc1, cc1plus, jc1, f771, etc.
+   Decode command args, then call compile_file.
+   Exit code is FATAL_EXIT_CODE if can't open files or if there were
+   any errors, or SUCCESS_EXIT_CODE if compilation succeeded.
+
+   It is not safe to call this function more than once.  */
+
+int
+toplev_main (argc, argv)
+     int argc;
+     char **argv;
+{
+  /* Initialization of GCC's environment, and diagnostics.  */
+  general_init (argv [0]);
+
+  /* Parse the options and do minimal processing; basically just
+     enough to default flags appropriately.  */
+  parse_options_and_default_flags (argc, argv);
+
+  /* Exit early if we can (e.g. -help).  */
+  if (!exit_after_options)
+    do_compile ();
 
   if (errorcount || sorrycount)
     return (FATAL_EXIT_CODE);