[PATCH] AC_C_BIGENDIAN: fix narrowing error

The following patch will fix AC_C_BIGENDIAN, for cross compiling, if 
narrowing is a fatal error in the used compiler.

error: narrowing conversion of ‘35283’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]
    { 0x89D3, 0xE3E3, 0x8593, 0x95C5, 0x89C4, 0x9581, 0 };
                                                        ^
error: narrowing conversion of ‘58339’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]
error: narrowing conversion of ‘34195’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]
error: narrowing conversion of ‘38341’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]
error: narrowing conversion of ‘35268’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]
error: narrowing conversion of ‘38273’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]
error: narrowing conversion of ‘49865’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]
    { 0xC2C9, 0xC785, 0x95C4, 0x8981, 0x95E2, 0xA8E2, 0 };
                                                        ^
error: narrowing conversion of ‘51077’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]
error: narrowing conversion of ‘38340’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]
error: narrowing conversion of ‘35201’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]
error: narrowing conversion of ‘38370’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]
error: narrowing conversion of ‘43234’ from ‘int’ to ‘short int’ inside 
{ } [-Wnarrowing]


--- a/lib/autoconf/c.m4
+++ b/lib/autoconf/c.m4
@@ -1655,9 +1655,9 @@ AC_DEFUN([AC_C_BIGENDIAN],
                 int use_ascii (int i) {
                   return ascii_mm[i] + ascii_ii[i];
                 }
-               short int ebcdic_ii[] =
+               unsigned short int ebcdic_ii[] =
                   { 0x89D3, 0xE3E3, 0x8593, 0x95C5, 0x89C4, 0x9581, 0 };
-               short int ebcdic_mm[] =
+               unsigned short int ebcdic_mm[] =
                   { 0xC2C9, 0xC785, 0x95C4, 0x8981, 0x95E2, 0xA8E2, 0 };
                 int use_ebcdic (int i) {
                   return ebcdic_mm[i] + ebcdic_ii[i];