Re: [RFA] Fix float argument passing in inferior function calls for ppc64

[Thiago Jung Bauermann <bauerman@br.ibm.com>]

On Wed, 2008-01-16 at 10:48 -0200, Thiago Jung Bauermann wrote:
> On Tue, 2008-01-15 at 18:54 -0200, Thiago Jung Bauermann wrote:
> > On Tue, 2008-01-15 at 21:48 +0100, Mark Kettenis wrote:
> > > > From: Thiago Jung Bauermann <bauerman@br.ibm.com>
> > > > Date: Tue, 15 Jan 2008 18:33:54 -0200
> > > > 
> > > > On Tue, 2008-01-15 at 15:39 -0200, Thiago Jung Bauermann wrote:
> > > > > > Is the first word used for anything in the new ABI?  If not, you could
> > > > > > support both ABIs by copying the value into both the first and the
> > > > > > second word.
> > > > > 
> > > > > Great idea. I will resend the patch using that approach then.
> > > > 
> > > > What about this patch?
> > > 
> > > Looks fine to me, but I have no means to test it.
> > 
> > This is tested in Linux/ppc64.
> > I'll check this version for regressions (I only regtested the previous
> > patch).
> 
> No regressions found. Ok to commit?

Ping?

I found an older Linux system here (from 2003 or 2004), and its GDB
implements the 1.7 ABI but still fails the test with many float
arguments, which means the 1.9 ABI is already in use.

Since current GDB CVS doesn't compile in that system anymore (actually
compilation failed in the bfd library), it also won't compile in an even
older system which would use the 1.7 ABI. Because of this, we shouldn't
worry too much about supporting the older ABI I think.

Still, my patch preserves compatibility with the 1.7 ABI while
implementing the 1.9 version. It's just that I wasn't able to test this
compatibility.
-- 
[]'s
Thiago Jung Bauermann
Software Engineer
IBM Linux Technology Center