Re: ARM long branch stubs: be8

[Christophe LYON <christophe dot lyon at st dot com>]

> Since I'm working on the stubs right now, I hope you don't mind that I
> checked this in for you.
No problem. Sorry for the timezone difference :-)

> The new tests pass both before and after;
> I added this additional patch with a big-endian test that fails after.
I don't understand: I have just tried and the 2 new tests pass here 
(arm-none-eabi).

> BE8 and little endian code have equivalent layout, even in Thumb; it's
> big-endian that is the odd duck out.
Sorry I had overlooked at this (we don't use this mode at all, here).

What confused me is that objdump -d dumps opcodes in a different order 
in the BE8 case. Maybe it could be enhanced to make use of the BE8 EABI 
flag?

Christophe.