Re: [PATCH] gas value truncation warning reports truncated values, doesn't look at signedness.

[Alan Modra]

On Mon, Jan 12, 2004 at 02:02:31AM +0100, Bart Samwel wrote:
> OK, I understand why it wasn't there yet. If you're up to ANSI C++ now, 
> a fix would be nice. :)

I'll put it on my TODO list.  There are many other places that need
this sort of fix..

> >I think you are confused over the rules for the type of an integer
> >constant.  Assuming 32 bit ints, the type of 0x7fffffff is "int", but
> >the type of 0x80000000 is "unsigned int".  Thus ~0x8000000 is
> >0x7fffffff, *not* 0xffffffff7fffffff as you claim.
> 
> Hmmm, then there is probably a bug in gas. The reason: why does the 
> warning happen on 0xffffffff7fffffff instead of 0x000000007fffffff as 
> the original value? Try using this code:
> 
> foo: .long ~(0x80000000)

Interesting.  The rules I was quoting were for ISO C90 rather than
gas.  One answer for the gas behavour is that gas is using 64 bit ints
when configured for a 64 bit target.

-- 
Alan Modra
IBM OzLabs - Linux Technology Centre