On Sat, Oct 30, 2004 at 11:18:42AM +0200, H.Merijn Brand wrote: > On Sat 30 Oct 2004 07:43, Ed Allen Smith <[EMAIL PROTECTED]> wrote: > > > > Oh, a couple bits of info: > > We've hit a point where it is wise to move to perl5-porters, since there's > more knowledge over there. > > The interrested should start here > http://www.mail-archive.com/[EMAIL PROTECTED]/msg00668.html > to read the full thread, minus first post which can be found here > http://www.mail-archive.com/[EMAIL PROTECTED]/msg20597.html
Having actually read the original message in my inbox and deleted it, I should have replied: you can already build -Dnoextensions=Encode and all tests pass. It's some of the other extensions that you can't do that with (yet - for most it's a bug) You might also find that ccache accelerates you build, and hence reduces you smoke times (if you're not already using it). Thinks like toke.c give most compilers indigestion, yet they don't change frequently, so they're an idea candidate for ccache. > > C++ 5.5.1 complier, returns "ok, ok" on a cc -n32 compile on a (n)32- > > machine; ditto on a cc -n32 compile on a 64-bit machine; ditto for cc -64; > > ditto for cc -32, for that matter. So that doesn't appear to be it. I think that the best thing to do at the moment is disable it in the hints file for the appropriate compiler/flags combinations, release 5.8.6 with this. Meanwhile work out what the real problem is in blead, commit it, and smoke it for a while until we're confident that it's stable (and that no-one has any late ideas for improving it) Nicholas Clark
