Re: dwarf2-frame.c read_reg problems, again ...

[Daniel Jacobowitz <drow@false.org>]

On Wed, Oct 31, 2007 at 07:37:03PM +0100, Ulrich Weigand wrote:
> So I don't mind making them unsigned again.  How would that work
> with the XML definitions?  Do you have to add an explicit type
> attribute everywhere?

That's right.  The default is "int", which is signed.

When I was designing the format I thought up a pile of clever ways to
make the format easier to write and more concise, like repeating
register sets.  Of course, they also make it harder to parse and
more complicated to document.  So I haven't implemented any of them
:-)

> I'd say targets like that should be able to define a proper conversion to
> builtin_type_void_data_ptr via convert_register or value_from_register
> for that specific register.
> 
> Alternatively, if builtin_type_void_data_ptr is in fact the wrong
> type to describe a CFA on some weird platform, we could allow the
> gdbarch to define the type to be used for this.

That's an interesting idea...

I don't have a helpful opinion about this particular case, I'm afraid.
I just don't want to break whatever Michael was working with, or
MIPS again (which requires sign extension here).

-- 
Daniel Jacobowitz
CodeSourcery