libtool-patches
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH] Removal of obsolete AmigaOS support


From: Dalibor Topic
Subject: Re: [PATCH] Removal of obsolete AmigaOS support
Date: Sun, 05 Oct 2003 23:24:39 +0200
User-agent: Mozilla/5.0 (X11; U; Linux i586; en-US; rv:1.4) Gecko/20030624

Marcus Comstedt wrote:
Scott James Remnant <address@hidden> writes:


On Sun, 2003-10-05 at 20:33, Marcus Comstedt wrote:


As a step in implementing support for the new AmigaOS 4.0 in GCC, I
noticed that there are some ancient remains of "amigaos" support in
the libtool configuration scripts.  These are intended for the now
defunct GeekGadgets project though, and not applicable to the new
AmigaOS.  In fact, they prevent libtool from working with AmigaOS
4.0.  Therefore, I'm hereby submitting a patch to remove this obsolete
and incompatible code.


How defunct is the project?  Is there still a significant userbase that
might want to run things compiled using libtool, or is it so dead and
defunct that nobody uses it anymore?


The project has not been maintained for two years, with the exception
of the GCC port.  When I spoke to the current maintainer of the
GeekGadgets GCC port (Gunther Nikl <address@hidden>), he had this to
say:

| Right, but then I already disabled this part in ltconfig because it didn't
| work for me and I believe this *ixlibrary is obsolete. Nobody is using it
| anymore.

So, I think that it can be assumed that there is no point in keeping
this code.

It worked well enough for building kaffe 1.0.7 (which is far from being a trivial app) on m68k-amigaos last time around. In fact, our amigaOS developer contributed a patch or two to libtool to make sure the latest libtool 1.5 runs well on amigaOS 3.9 on m68k machines. Those patches are in libtool CVS now.

The major problem right now in running libtool 1.5 plus amigaos patches from CVS on m68k based machines (amigaos, or not) is the very inefficent check for maximum allowed variable length, that can take up to an hour, or more.

Does the GCC stuff use the amigaos* configuration name as well?  If not,
I don't see how it's incompatible, if it does then surely an extra case
statement would distinguish the two?


Well, they can be distinguished by the CPU architecture, since the new
OS is currently PPC only, and the old stuff is for m68k.

sounds like the way to go for me.

cheers,
dalibor topic





reply via email to

[Prev in Thread] Current Thread [Next in Thread]