On 1/4/06, Tushar Teredesai <[EMAIL PROTECTED]> wrote: > On 1/3/06, Dan Nicholson <[EMAIL PROTECTED]> wrote: > > For evidence, here's a diff of the build logs between iter1 and iter2 for > > bison: > > > > --- iter1/bison-2.1.log 2005-12-29 11:50:40.000000000 -0800 > > +++ iter2/bison-2.1.log 2005-12-29 13:28:54.000000000 -0800 > > @@ -79,8 +79,7 @@ > > checking for yywrap in -lfl... yes > > checking lex output file root... lex.yy > > checking whether yytext is a pointer... yes > > -checking for bison... no > > -checking for byacc... no > > +checking for bison... bison -y > > checking for ranlib... ranlib > > checking for gm4... no > > checking for gnum4... no > > Some of the configure checks are not useful. That is the results don't > get used during compilation and hence won't make a difference in ICA. > There is a similar check in binutils.
OK, but in this case it was useful. Putting bison in Ch.5, and moving flex before bison in Ch. 6 has cleared up the ICA differences. Later I'll actually look at the bison source a bit to confirm this, but I'm pretty sure there's a circular dependency here. flex uses bison, and bison seems to use these features in flex if they exist. If anyone understands how these packages depend on each other, feel free to chime in. I realize that configure can produce a lot of red herrings, but it seemed like a reasonable angle to at least pursue. I made a similar change due to e2fsprogs and gettext that I think I'll reverse. -- Dan -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
