This is the mail archive of the
gcc-help@gcc.gnu.org
mailing list for the GCC project.
gnu_parallel STL not vectorized?
- From: Tim Prince <n8tm at aol dot com>
- To: gcc-help <gcc-help at gcc dot gnu dot org>
- Date: Thu, 23 Apr 2015 10:11:29 -0400
- Subject: gnu_parallel STL not vectorized?
- Authentication-results: sourceware.org; auth=none
- Reply-to: tprince at computer dot org
>From the looks of gcc 6.0, it appears that gnu_parallel STL is slated to
become part of the release.
Am I right in believing that gnu_parallel STL code is not simd
vectorized? Will there be an option to change this?
PR60117 may be related in that it is a case where simd reduction is
avoided with OpenMP. It's still marked unconfirmed even though I see
that a useful discussion was appended over a year ago.
I am able to build and run trunk on cygwin64 by papering over the
winnt.c bootstrap failures (PR65835) which I guess have to do with the
compiler checking validity of const char * usage. I don't understand
the code well enough to propose a serious patch.
--
Tim Prince