All,
my sincerest apologies if I am missing something obvious, but
these defines from "gas/config/aout_gnu.h" sure look fishy.
the big-endian defines all march down in value from 0x80 to 0x02,
to match bits numbers 0 to 6 of the unsigned char r_bits[1]
but the little-endian values aren't consistently increasing or
decreasing,
it looks like some cut-n-pasting took place without some necessary
editing...
#define RELOC_STD_BITS_PCREL_BIG 0x80
#define RELOC_STD_BITS_PCREL_LITTLE 0x01
#define RELOC_STD_BITS_LENGTH_BIG 0x60
#define RELOC_STD_BITS_LENGTH_SH_BIG 5
#define RELOC_STD_BITS_LENGTH_LITTLE 0x06
#define RELOC_STD_BITS_LENGTH_SH_LITTLE 1
#define RELOC_STD_BITS_EXTERN_BIG 0x10
#define RELOC_STD_BITS_EXTERN_LITTLE 0x08
#define RELOC_STD_BITS_BASEREL_BIG 0x08
#define RELOC_STD_BITS_BASEREL_LITTLE 0x08 <<=== shouldn't
this be 0x10
#define RELOC_STD_BITS_JMPTABLE_BIG 0x04
#define RELOC_STD_BITS_JMPTABLE_LITTLE 0x04 <<=== shouldn't
this be 0x20
#define RELOC_STD_BITS_RELATIVE_BIG 0x02
#define RELOC_STD_BITS_RELATIVE_LITTLE 0x02 <<=== shouldn't
this be 0x40
so for example without the suggested fixes we have little-endian
conflicts
_EXTERN_LITTLE has the exact same value 0x08 as _BASEREL_LITTLE
and _LENGTH_LITTLE value 0x06 gets overlapped by _JMPTABLE_LITTLE
0x04
and _RELATIVE_LITTLE 0x02.
sincerely,
Peter Lawrence.
_______________________________________________
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
