Re: drag-and-drop failures

[Dave Love]

"Jan D." <address@hidden> writes:

> I have checked in a change in Fbyteorder so it behaves on 64 bit
> also.

I think it did actually work, because unsigned is 32-bit on all the
relevant 64-bit systems.  unsigned long is 64-bit.

> The reason for a runtime check was that is the same type of check that
> X11 uses, and I didn't know about WORDS_BIG_ENDIAN...  Wasn't there
> a processor that could be both little and big endian?

Yes.  There are Emacs configurations for both byte sexes of MIPS, for
instance, but I doubt a given OS would run both ways round, and it
seems highly unlikely it could be a runtime issue for a given binary.
For what it's worth, at least PowerPC, (some) SPARC, ARM, SH and
Itanic can also be BE or LE (e.g. the GCC submodel options).