[Aboriginal] undefined reference to `finite'

David Seikel onefang at gmail.com
Tue Apr 12 09:33:48 PDT 2011


On Fri, 8 Apr 2011 01:17:19 +1000 David Seikel <onefang at gmail.com>
wrote:

> On Thu, 7 Apr 2011 23:21:34 +1000 David Seikel <onefang at gmail.com>
> wrote:
> 
> > On Tue, 5 Apr 2011 06:35:02 -0500 Rob Landley
> > <rlandley at parallels.com> wrote:
> > 
> > > On 04/05/2011 05:12 AM, David Seikel wrote:
> > > > I'm very tired, and slightly ill, and I can't figure this out.
> > > > 
> > > > One of the packages from the UI library I want to use for this
> > > > project uses a function called finite(double).  As far as I can
> > > > tell, uClibc should supply that function, since we have
> > > > DO_C99_MATH=y in its baseconfig. It's not being found though.
> > > > 
> > > > It's all buried in a forest of #ifdef's that my tired mind is in
> > > > no condition to sort out.
> > > > 
> > > > I'm only two packages away for having all my needed libraries
> > > > installed before I can start cutting code.  Time to sleep.
> > > 
> > > Hmmm...  Posibly a question for the uClibc mailing list?
> > > 
> > > You didn't say which target you were using, so I'll poke at
> > > x86_64:
> > 
> > I'm sticking with doing my builds for 486, since it's still possible
> > that those tiny 486 based boards I looked at last year may be used.
> > If they are not used, then 486 code will at least work fine on
> > whatever we move to.  Any board we move to will be a lot faster
> > anyway, and I've already mentioned that speed is not important for
> > this project.
> > 
> > It appears that finite() is obsolete, and replaced with an
> > isfinite() macro.  A few things I found in my research pointed to
> > that, including the man pages on Ubuntu, but the clincher was this -
> > 
> > http://www.mail-archive.com/uclibc@uclibc.org/msg03404.html
> > 
> > I'll poke the UI library development list about changing that,
> > though the source I'm using is a stable release tarball, but they
> > are thinking of cutting another release soon.  Wont be a major
> > change, I think they only use finite() once.

The UI library upstream have put in a suitable fix now.

-- 
A big old stinking pile of genius that no one wants
coz there are too many silver coated monkeys in the world.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://lists.landley.net/pipermail/aboriginal-landley.net/attachments/20110413/015129b6/attachment-0003.pgp>


More information about the Aboriginal mailing list