amd64-build problems//now problem with TCL

Ken Moffat ken at
Sat Jan 5 12:59:15 PST 2008

On Sat, Jan 05, 2008 at 09:11:23PM +0100, at wrote:
> lieee -lm   -Wl,-rpath,/tools/lib
> fixstrtod.o: In function `fixstrtod':
> fixstrtod.c:(.text+0x0): multiple definition of `fixstrtod'
> strtod.o:strtod.c:(.text+0x0): first defined here
> collect2: ld returned 1 exit status
> make: *** [] Error 1
> Installing private header files
 Thanks for cutting and pasting.  There are occasional reports of
this, and last April Tijnema suggested:

> try running:
> export ac_cv_func_strtod=yes
> and then run configure and make again.

 I don't know if it worked, but it's the best suggestion I can find.
That seems to be based on
 which has a further additional export as sometimes needed.

 All in all, it's a little disconcerting, because nobody has
reported this recently, and it is unclear in what circumstances it
(or 'they') is/are needed.  Wearing one of my other hats, I'm happy
to throw the proverbial kitchen-sink full of ac_cv and similar
spells at a _cross_-compiler, albeit I always put them in
config.cache and pass a pointer to that.  Shouldn't usually be
necessary, but maybe there's something about the host distro.  (If
it comes to that, I've no idea how my own 'pure64' builds will cope
with a native build, so it might be something that needs to be added
in moving to x86_64.)

 Hang on, I don't think changes for x86_64 made it out of the 'jh'
branch.  If you are really following the svn book, you're going to
have some disappointment if you try to build grub from a 64-bit
toolchain.  As far as I know, the jh branch isn't publically rendered
to html anywhere (so it's only available via subversion), the last
update to it was in October, and it doesn't yet include bootloaders.

das eine Mal als Tragödie, das andere Mal als Farce

More information about the lfs-support mailing list