I fear the idea that valid bugs may be closed do to a -march=i586. release media should not have to be tuned to i386. perhaps thes older machines shouldn't be a priority, but that doesn't mean they should become completely unsupported. if a general move to i686 is desired perhaps the archs should split x86 and i686 or some such. and applications that are unable to be supported on < i686 be removed from the x86 tree.
On 10/11/06, Duncan <[EMAIL PROTECTED]> wrote:
Paul de Vrieze <[EMAIL PROTECTED]> posted [EMAIL PROTECTED], excerpted below, on Tue, 10 Oct 2006 16:46:05 +0200: > A couple of years ago (when we were still using gcc-2.95 I used to run > gentoo on my server machine which was a pentium-60 (with fdiv bug). While > it took a while to compile the bigger packages it was certainly workable. > I did it because I didn't have a better machine, not to be able to say I > did it. Well yes, except that I'd guess that was a bit more than a couple of years ago (I've been on Gentoo since 2004.0/2004.1, and IIRC it was gcc-3.3 then, so 2.95 would have been what, at least three years ago??). That means the archs are a third(-ish) of a decade further out of date than they were then. That's a significant amount of time in computer terms. Anyway, not supported doesn't mean can't do it. As I suggested in a different reply, it could and would likely still be done, just as Gentoo based systems are run on all sorts of stuff according to embedded, and in fact they may choose to continue some support, as I believe pentium-class embedded is quite popular. Not supported just means less frequent install media or bootstrapping from other distributions instead of Gentoo install media, and that bugs can be closed if desired and appropriate, based on that alone. -- Duncan - List replies preferred. No HTML msgs. "Every nonfree program has a lord, a master -- and if you use the program, he is your master." Richard Stallman -- gentoo-dev@gentoo.org mailing list
-- gentoo-dev@gentoo.org mailing list
