This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: how to enable vector intrinsic functions in fortran frontend
- From: "Andrew Pinski" <pinskia at gmail dot com>
- To: "Sa Liu" <SALIU at de dot ibm dot com>
- Cc: "Tobias Burnus" <burnus at net-b dot de>, fortran at gcc dot gnu dot org, "Ulrich Weigand" <Ulrich dot Weigand at de dot ibm dot com>, "Tobias SchlÃter" <tobias dot schlueter at physik dot uni-muenchen dot de>, "Steve Kargl" <sgk at troutmask dot apl dot washington dot edu>
- Date: Wed, 17 Oct 2007 11:08:54 -0700
- Subject: Re: how to enable vector intrinsic functions in fortran frontend
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; bh=msl3E0Ts4Evbj7YBmjYRRRf5QutPp9S8CPLoKvlD/qQ=; b=LjXs8iZ9uwP3P47kupDDhOAwEFLYnkj9MR5rJ6l8j+u89UNb5tuFqLVKjXLj+eg1mjyXyzhz2HLBKBLaKG2GIgBzvPP5Wc0ZVU0vv6GY+DONIcAe1W84kHBKQ0nZBv2cCHg1IM3lqY1yd/tSOYBqNsxaAFyGI0Qs+CnHhjEBUqY=
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=lTnivRgslP6HrZciyrWttS9q2RNyGr+bPDzOIJ7v4Jq8JlCQMT4Z+Wyi3qfP0sj7l9jZx8aRSY+M9TvpdrGb/9pNDZMALi0I7Hrwj8YHzaDREcPmsebY+N0XrGibS4qpYQAhS5h0g4bBQUHBZ4tqlOUq1aUBDdokUqIcbTrcB6Q=
- References: <4713B9ED.8070504@net-b.de> <OFC22126C3.7B40A434-ONC1257377.0055F876-C1257377.005D9DCD@de.ibm.com>
On 10/17/07, Sa Liu <SALIU@de.ibm.com> wrote:
May I suggest that the VECTOR "kind" takes an optional argument which
is exactly the same as the vector_size attribute argument. This
allows for generic vector support in gfortran and uses the already
implemented generic vector support inside GCC?
The idea would allow:
vector(Integer, 16) :: a, b, c
a = b + c
end
so that would be constent with the generic vector support in the C/C++
front-ends.
Maybe it is better to implement this first and then implement the
VMX/SPU specific parts.
Thanks,
Andrew Pinski