| delorie.com/archives/browse.cgi | search |
On Fri, Sep 08, 2000 at 09:58:30AM -0400, Larry Hall (RFK Partners, Inc) wrote: >Date: Fri, 08 Sep 2000 09:58:30 -0400 >To: cygwin developers <cygwin-developers AT sourceware DOT cygnus DOT com> >From: "Larry Hall (RFK Partners, Inc)" <lhall AT rfk DOT com> >Subject: Re: -lc and -lm >In-Reply-To: <20000908123517 DOT 25018 DOT qmail AT web124 DOT yahoomail DOT com> > >At 08:35 AM 9/8/2000, Earnie Boyd wrote: >>Back in B18 when I started using Cygwin these libraries were stub libraries. >>Is there a reason that they shouldn't be stub libraries instead of symlinks to >>cygwin runtime? > >Good question. I remember a discussion on the topic of exactly what form >these libraries should take in Cygwin back a long time ago. I believe it >was Mumit who suggested that these libraries could (and should) be symlinks >(I may not be remembering this correctly.) Anyway, its my impression that >having libm.a and libc.a be symlinks to libcygwin.a is sufficiently >problematic that it makes sense to explore other options. > >There. Now that we have my opinion, someone can go ahead and fix the >problem!;-) What does "problematic" mean? The reason for making them something other than stubs is that some packages search for symbols there. IF they are empty libraries there won't be any symbols to search for. cgf
| webmaster | delorie software privacy |
| Copyright © 2019 by DJ Delorie | Updated Jul 2019 |