On Fri, 7 Jun 2024, Tobias Burnus wrote: >> + <li id="openmp"><a href="https://gcc.gnu.org/projects/gomp/";>OpenMP</a> >> Can you please make this a relative link, i.e. "../projects/gomp/"? > Good point. I thought such links should be absolute because of (www.)GNU.org, > i.e. > > https://www.gnu.org/software/gcc/releases.html
We only need to use absolutely links for material only available on gcc.gnu.org such as Bugzilla, the Wiki, or /onlinedocs. Everyone directly unter wwwdocs/htdocs can be relative. > GNU.org does not have the documentation, but going to > https://www.gnu.org/software/gcc/onlinedocs/ or a subpage redirects (302 > temporary redirect) to the GCC website. Likewise for '../git' but for > '../wiki' it has a HTTP 404 not found; fortunately, ../wiki/ works. > > I think there are plenty of links which could be relative ones but are > absolute ones. The original trigger to be careful and provide absolute links was https://www.gnu.org/software/gcc/ which was established early in the days of GCC 2.95 after the egcs/gcc/GCC reconciliation. This is also useful in case anyone wants to use these pages locally. > Looks like a janitorial task to fix the absolute links, possibly > excluding those with /git, /onlinedocs, /wiki – or assuming that the > main page is GCC.gnu.org, relying on the redirects. It's on my list. A first quick check indicates there isn't much to do, though. :-) > o In any case, those links are probably broken on GNU.org: > > htdocs/gcc-14/porting_to.html:<a > href="/onlinedocs/gcc-14.1.0/gcc/Diagnostic-Pragmas.html"><code>#pragma GCC > diagnostic warning</code></a> > > htdocs/gcc-5/changes.html: <li>A <a > href="/onlinedocs/libstdc++/manual/using_dual_abi.html">Dual Yes, those are definitely in need of being fixed. I'll do so in a minute. Thanks for pointing out the two. >> + loop-transformation constructs are now supported. >> I'm thinking "loop transformation" in English? Or is this a specific term >> from the standard? > Loop transformation happens at the end. But e.g "(#pragma omp) unroll > full" is a directive and, e.g. > > #pragma omp unroll partial(2) > > for (int i=0; i < n; i++) > > a[i] = 5; > > is a construct (= directive + structured block (if any) + end directive > (if any)). I believe there was a misunderstanding and I wasn't clear enough: I was wondering whether instead of "loop-transformation" the patch should have "loop transformation". In your response you use the version without dash, so I guess we agree? :-) Cheers, Gerald
