Fortran: Cast arguments of <ctype.h> functions to unsigned char

Functions from <ctype.h> should only be called on values that can be
represented by unsigned char. On targets where char is a signed type,
some of libgfortran calls have undefined behaviour.

The solution is to cast the argument to unsigned char type. I’ve defined
macros in libgfortran.h to do so, to retain legibility of the library
code.

PR libfortran/95177

libgfortran/ChangeLog

	* libgfortran.h: include ctype.h, provide safe macros.
	* io/format.c: use safe macros.
	* io/list_read.c: use safe macros.
	* io/read.c: use safe macros.
	* io/write.c: use safe macros.
	* runtime/environ.c: use safe macros.

6 files changed, 37 insertions, 27 deletions

diff --git a/libgfortran/io/format.c b/libgfortran/io/format.c
index 56f8dbd..927e378 100644
--- a/libgfortran/io/format.c
+++ b/libgfortran/io/format.c
@@ -29,7 +29,6 @@ see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
 
 #include "io.h"
 #include "format.h"
-#include <ctype.h>
 #include <string.h>
 
 
@@ -193,7 +192,7 @@ next_char (format_data *fmt, int literal)
 	return -1;
 
       fmt->format_string_len--;
-      c = toupper (*fmt->format_string++);
+      c = safe_toupper (*fmt->format_string++);
       fmt->error_element = c;
     }
   while ((c == ' ' || c == '\t') && !literal);
@@ -328,7 +327,7 @@ format_lex (format_data *fmt)
 
     case '+':
       c = next_char (fmt, 0);
-      if (!isdigit (c))
+      if (!safe_isdigit (c))
 	{
 	  token = FMT_UNKNOWN;
 	  break;
@@ -339,7 +338,7 @@ format_lex (format_data *fmt)
       for (;;)
 	{
 	  c = next_char (fmt, 0);
-	  if (!isdigit (c))
+	  if (!safe_isdigit (c))
 	    break;
 
 	  fmt->value = 10 * fmt->value + c - '0';
@@ -367,7 +366,7 @@ format_lex (format_data *fmt)
       for (;;)
 	{
 	  c = next_char (fmt, 0);
-	  if (!isdigit (c))
+	  if (!safe_isdigit (c))
 	    break;
 
 	  fmt->value = 10 * fmt->value + c - '0';
diff --git a/libgfortran/io/list_read.c b/libgfortran/io/list_read.c
index 8cc7ddb..f902ee4 100644
--- a/libgfortran/io/list_read.c
+++ b/libgfortran/io/list_read.c
@@ -29,7 +29,6 @@ see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
 #include "fbuf.h"
 #include "unix.h"
 #include <string.h>
-#include <ctype.h>
 
 typedef unsigned char uchar;
 
@@ -811,7 +810,7 @@ read_logical (st_parameter_dt *dtp, int length)
   if (parse_repeat (dtp))
     return;
 
-  c = tolower (next_char (dtp));
+  c = safe_tolower (next_char (dtp));
   l_push_char (dtp, c);
   switch (c)
     {
@@ -837,7 +836,7 @@ read_logical (st_parameter_dt *dtp, int length)
       break;
 
     case '.':
-      c = tolower (next_char (dtp));
+      c = safe_tolower (next_char (dtp));
       switch (c)
 	{
 	  case 't':
@@ -1052,7 +1051,7 @@ read_integer (st_parameter_dt *dtp, int length)
     }
 
  get_integer:
-  if (!isdigit (c))
+  if (!safe_isdigit (c))
     goto bad_integer;
   push_char (dtp, c);
 
@@ -1303,7 +1302,7 @@ parse_real (st_parameter_dt *dtp, void *buffer, int length)
   if (c == ',' && dtp->u.p.current_unit->decimal_status == DECIMAL_COMMA)
     c = '.';
 
-  if (!isdigit (c) && c != '.')
+  if (!safe_isdigit (c) && c != '.')
     {
       if (c == 'i' || c == 'I' || c == 'n' || c == 'N')
 	goto inf_nan;
@@ -1377,7 +1376,7 @@ parse_real (st_parameter_dt *dtp, void *buffer, int length)
     }
 
  exp2:
-  if (!isdigit (c))
+  if (!safe_isdigit (c))
     {
       /* Extension: allow default exponent of 0 when omitted.  */
       if (dtp->common.flags & IOPARM_DT_DEC_EXT)
@@ -1748,7 +1747,7 @@ read_real (st_parameter_dt *dtp, void *dest, int length)
   if (c == ',' && dtp->u.p.current_unit->decimal_status == DECIMAL_COMMA)
     c = '.';
 
-  if (!isdigit (c) && c != '.')
+  if (!safe_isdigit (c) && c != '.')
     {
       if (c == 'i' || c == 'I' || c == 'n' || c == 'N')
 	goto inf_nan;
@@ -1828,7 +1827,7 @@ read_real (st_parameter_dt *dtp, void *dest, int length)
     }
 
  exp2:
-  if (!isdigit (c))
+  if (!safe_isdigit (c))
     {
       /* Extension: allow default exponent of 0 when omitted.  */
       if (dtp->common.flags & IOPARM_DT_DEC_EXT)
@@ -2757,7 +2756,7 @@ nml_match_name (st_parameter_dt *dtp, const char *name, index_type len)
   for (i = 0; i < len; i++)
     {
       c = next_char (dtp);
-      if (c == EOF || (tolower (c) != tolower (name[i])))
+      if (c == EOF || (safe_tolower (c) != safe_tolower (name[i])))
 	{
 	  dtp->u.p.nml_read_error = 1;
 	  break;
@@ -3286,7 +3285,7 @@ get_name:
   do
     {
       if (!is_separator (c))
-	push_char_default (dtp, tolower(c));
+	push_char_default (dtp, safe_tolower(c));
       if ((c = next_char (dtp)) == EOF)
 	goto nml_err_ret;
     }
diff --git a/libgfortran/io/read.c b/libgfortran/io/read.c
index 7515d91..7b3f137 100644
--- a/libgfortran/io/read.c
+++ b/libgfortran/io/read.c
@@ -28,7 +28,6 @@ see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
 #include "format.h"
 #include "unix.h"
 #include <string.h>
-#include <ctype.h>
 #include <assert.h>
 #include "async.h"
 
@@ -959,7 +958,7 @@ read_f (st_parameter_dt *dtp, const fnode *f, char *dest, int length)
 	 between "NaN" and the optional perenthesis is not permitted.  */
       while (w > 0)
 	{
-	  *out = tolower (*p);
+	  *out = safe_tolower (*p);
 	  switch (*p)
 	    {
 	    case ' ':
@@ -981,7 +980,7 @@ read_f (st_parameter_dt *dtp, const fnode *f, char *dest, int length)
 		goto bad_float;
 	      break;
 	    default:
-	      if (!isalnum (*out))
+	      if (!safe_isalnum (*out))
 		goto bad_float;
 	    }
 	  --w;
@@ -1109,7 +1108,7 @@ exponent:
 
   if (dtp->u.p.blank_status == BLANK_UNSPECIFIED)
     {
-      while (w > 0 && isdigit (*p))
+      while (w > 0 && safe_isdigit (*p))
 	{
 	  exponent *= 10;
 	  exponent += *p - '0';
@@ -1137,7 +1136,7 @@ exponent:
 	      else
 		assert (dtp->u.p.blank_status == BLANK_NULL);
 	    }
-	  else if (!isdigit (*p))
+	  else if (!safe_isdigit (*p))
 	    goto bad_float;
 	  else
 	    {
diff --git a/libgfortran/io/write.c b/libgfortran/io/write.c
index 278cd47..b9e9284 100644
--- a/libgfortran/io/write.c
+++ b/libgfortran/io/write.c
@@ -30,7 +30,6 @@ see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
 #include "unix.h"
 #include <assert.h>
 #include <string.h>
-#include <ctype.h>
 
 #define star_fill(p, n) memset(p, '*', n)
 
@@ -2101,14 +2100,14 @@ nml_write_obj (st_parameter_dt *dtp, namelist_info *obj, index_type offset,
 	  base_name_len = strlen (base_name);
 	  for (dim_i = 0; dim_i < base_name_len; dim_i++)
             {
-	      cup = toupper ((int) base_name[dim_i]);
+	      cup = safe_toupper (base_name[dim_i]);
 	      write_character (dtp, &cup, 1, 1, NODELIM);
             }
 	}
       clen = strlen (obj->var_name);
       for (dim_i = len; dim_i < clen; dim_i++)
 	{
-	  cup = toupper ((int) obj->var_name[dim_i]);
+	  cup = safe_toupper (obj->var_name[dim_i]);
 	  if (cup == '+')
 	    cup = '%';
 	  write_character (dtp, &cup, 1, 1, NODELIM);
@@ -2426,7 +2425,7 @@ namelist_write (st_parameter_dt *dtp)
   /* Write namelist name in upper case - f95 std.  */
   for (gfc_charlen_type i = 0; i < dtp->namelist_name_len; i++ )
     {
-      c = toupper ((int) dtp->namelist_name[i]);
+      c = safe_toupper (dtp->namelist_name[i]);
       write_character (dtp, &c, 1 ,1, NODELIM);
     }
 
diff --git a/libgfortran/libgfortran.h b/libgfortran/libgfortran.h
index 285c36a..93e3591 100644
--- a/libgfortran/libgfortran.h
+++ b/libgfortran/libgfortran.h
@@ -39,6 +39,7 @@ see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
 /* config.h MUST be first because it can affect system headers.  */
 #include "config.h"
 
+#include <ctype.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <stddef.h>
@@ -103,6 +104,20 @@ typedef off_t gfc_offset;
 #endif
 
 
+/* These functions from <ctype.h> should only be used on values that can be
+   represented as unsigned char, otherwise the behavior is undefined.
+   Some targets have a char type that is signed, so we cast the argument
+   to unsigned char. See:
+     https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95177
+     https://wiki.sei.cmu.edu/confluence/x/BNcxBQ
+ */
+
+#define safe_isalnum(x) isalnum((unsigned char) (x))
+#define safe_isdigit(x) isdigit((unsigned char) (x))
+#define safe_tolower(x) tolower((unsigned char) (x))
+#define safe_toupper(x) toupper((unsigned char) (x))
+
+
 /* The following macros can be used to annotate conditions which are likely or
    unlikely to be true.  Avoid using them when a condition is only slightly
    more likely/less unlikely than average to avoid the performance penalties of
diff --git a/libgfortran/runtime/environ.c b/libgfortran/runtime/environ.c
index fe16c08..ce408cf 100644
--- a/libgfortran/runtime/environ.c
+++ b/libgfortran/runtime/environ.c
@@ -26,7 +26,6 @@ see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
 
 #include <string.h>
 #include <strings.h>
-#include <ctype.h>
 
 #ifdef HAVE_UNISTD_H
 #include <unistd.h>
@@ -91,7 +90,7 @@ init_integer (variable * v)
     return;
 
   for (q = p; *q; q++)
-    if (!isdigit (*q) && (p != q || *q != '-'))
+    if (!safe_isdigit (*q) && (p != q || *q != '-'))
       return;
 
   *v->var = atoi (p);
@@ -344,7 +343,7 @@ static int
 match_integer (void)
 {
   unit_num = 0;
-  while (isdigit (*p))
+  while (safe_isdigit (*p))
     unit_num = unit_num * 10 + (*p++ - '0');
   return INTEGER;
 }