Re: AAPCS compatibility

[Richard Earnshaw <richard.earnshaw%buzzard.freeserve.co.uk@localhost>]

The major difference is alignment of 64-bit types, which in AAPCS is
64-bit, versus 32-bit in our ABI.

I think this might be a good time to look at making the switch, there
are a number of other benefits from doing so, such as access to a larger
range of compilers, better support for C++ exceptions, etc.

R.

On Sat, 2008-04-05 at 12:42 +0100, Chris Gilbert wrote:
> Hi,
> 
> I was wondering if anyone has tracked how close our ARM ABI is to AAPCS?
> 
> My reason for thinking about it is that should we shift to a 64bit
> time_t it would probably cause a libc major version bump, so it would
> seem sensible to do any other ABI changes then.
> 
> The latest AAPCS is available from:
> http://infocenter.arm.com/help/topic/com.arm.doc.ihi0042a/IHI0042A_aapcs.pdf
> (actually searching on infocenter.arm.com for ABI and r2.06 turns up the
> relevant docs covering elf and dwarf etc)
> 
> Looking through the AAPCS the only obvious difference is that size_t
> should is unsigned int in AAPCS, we say unsigned long int.
> 
> I believe that someone with more knowledge of the compiler would be able
> to say how close we actually are.
> 
> I thought I'd raise this in case others hadn't noticed the time_t branch.
> 
> Thanks,
> Chris
> 
>