This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: another ia64 (and maybe generic) vector mode issue
- From: Richard Henderson <rth at redhat dot com>
- To: Jan Beulich <JBeulich at novell dot com>
- Cc: gcc at gcc dot gnu dot org
- Date: Wed, 5 Jan 2005 04:48:00 -0800
- Subject: Re: another ia64 (and maybe generic) vector mode issue
- References: <s1dbd338.036@emea1-mh.id2.novell.com>
On Wed, Jan 05, 2005 at 12:44:54PM +0100, Jan Beulich wrote:
> So it is just on oversight/ommission that this isn't done so in
> ia64.c?
Yes.
> They pass them in integer registers, and also return them there.
Eh. 6 cycles of latency moving between fp and int registers,
but I guess there's worse things.
> They're not supporting anything but (in gcc notation) V4SFmode at
> present, though. I opened an issue with them to clarify whether they
> consider their compiler behavior ABI compliant.
Good. V2DF is also of interest, because of emmintrin.h. I'm going
to presume that V2SF should be passed as half of whatever answer you
get for V4SF.
r~