This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: [Bug target/16872] invalid use of MMX instructions
- From: "Jan Beulich" <JBeulich at novell dot com>
- To: <pinskia at gcc dot gnu dot org>
- Cc: <gcc-bugs at gcc dot gnu dot org>
- Date: Thu, 05 Aug 2004 18:08:13 +0200
- Subject: Re: [Bug target/16872] invalid use of MMX instructions
>You are explicately using the vector registers with vector_size so ...
Not really, I'm just using vector types. When there are no vector
registers, the compiler still accepts this and emulates it by using
scalar operations (if what you say is true, then I can't see the point
of it doing so). Thus to me these are two different things, and thus I
continue to believe these are two different problems. (To me, the vector
support offered by gcc goes beyond the SIMD support in e.g. the Intel
compiler, because regardless of architecture certain vector types are
defined, and they can be used with normal operations, rather than only
through intrinsics. [Generally, however, the vector mode support seems
to lack some fundamental design.])
Jan