Re: [doc,libgomp] Updates for www.openacc.org URLs

[Thomas Schwinge <thomas at codesourcery dot com>]

Hi Gerald!

On Fri, 8 Sep 2017 14:27:30 +0200, Gerald Pfeifer <gerald@pfeifer.com> wrote:
> We were nearly down to around a dozen of problematic links on 
> gcc.gnu.org and our online docs, when the following popped up: 
> www.openacc.org switched to https

I can't tell what is "problematic" here, given that a redirect is being
done (and most likely will be "forever")?

    $ curl -v http://www.openacc.org/
    *   Trying 23.185.0.1...
    * Connected to www.openacc.org (23.185.0.1) port 80 (#0)
    > GET / HTTP/1.1
    > Host: www.openacc.org
    > User-Agent: curl/7.50.1
    > Accept: */*
    >
    < HTTP/1.1 301 Moved Permanently
    < Content-Type: text/html; charset=UTF-8
    < Location: https://www.openacc.org/
    [...]

> (and does not require a trailing 
> slash).

I prefer the form with a trailing slash, but I'm certainly no going to
lose any sleep over that.  ;-)

> I committed the patch below, but am wondering whether we really,
> really, really need as many instances of one and the same URL in 
> one section of the documentation?

Well, that is just OpenACC doing what OpenMP had done before.

> Can we simplify this somehow?

My preference actually is to just remove all this boilerplate stuff,
which, as far as I can tell, is just text copied from the respective
specifications anyway.  That is, why do we highlight the runtime library
APIs here, when we nowhere document the directives in much detail, which
are actually way more important, usually?  And if GCC claims to implement
these standards, then the user should rely on these standards (aside from
implementation-defined aspects, of course), just like we do for other
programming language standards?


Grüße
 Thomas