[RFC] WORDS_BIG_ENDIAN != BYTES_BIG_ENDIAN messes up subreg

[Richard Henderson]

On 10/12/2010 02:31 PM, Paul Koning wrote:
> I just ran into an issue when BYTES_BIG_ENDIAN differs from
> WORDS_BIG_ENDIAN, which apparently at this point applies only to
> pdp11.

Yep, it would do.  Can you point to an exemplar -- preferably with
a bug report for tracking -- that hits this?

I have a feeling that while your patch works for HImode, it'll 
fail for a QImode subreg of an SImode value.  Which means that
we'll need some other adjustment in there instead.


r~