Guy Harris <gharris@xxxxxxxxx> writes:
> On Fri, Apr 12, 2002 at 08:45:21AM +0200, Falk Hayn wrote:
> > /usr/ccs/bin/ld: Unsatisfied symbols:
> > fcstab_32 (code)
> > table_v5 (code)
> > PGM_OPT_PARITY_PRM_OND (code)
> > PGM_NAK_SZ (code)
> > PGM_OPT_INVALID (code)
> > PGM_ACK_SZ (code)
> > PGM_OPT_PARITY_PRM_PRO (code)
> > PGM_SPM_SZ (code)
> > PGM_DATA_HDR_SZ (code)
> > fcstab_16 (code)
>
> These are just "const" symbols defined in various files; there's
> probably some mysterious compiler or linker problem, or a problem of
> some sort with libtool.
>
> Unfortunately, I have no idea what the problem might be.
This is going to a little vague since I no longer have access to the
original bug report or an HP compiler, but the combination of "HP",
"const" and "unsatisfied symbols" reminds me of a problem I saw
several years ago.
What we discovered was that under some circumstances the HP linker
would not bother to link in const objects. To the best of my
recollection we had a file containing a number of const objects, but
unless we also included a junk function in the file and ensured that
it got called somewhere the const objects never got linked in.
Not sure if your problem is related or not, but it sounded similar
enough that I thought I would mention it.
-Andrew
>
> > /usr/ccs/bin/ld: (Warning) Linker features were used that may not be
> > supported in future releases. The +vallcompatwarnings option can be used to display
> > more details, and the ld(1) man page contains additional information. This
> > warning can be suppressed with the +vnocompatwarnings option.
>
> That's probably another libtool problem; it may be passing some
> command-line option to the linker that HP are considering removing in
> the future.
>
> You might check the page at
>
> http://www.gnu.org/software/libtool/libtool.html
>
> and see if there's anything about problems like that (either with
> Ethereal or with other programs - often build problems with Ethereal are
> just generic problems with build tools, and show up with other programs
> as well) in any of the mailing lists.
>
> _______________________________________________
> Ethereal-dev mailing list
> Ethereal-dev@xxxxxxxxxxxx
> http://www.ethereal.com/mailman/listinfo/ethereal-dev
>
--
-Andrew Feren
Cetacean Networks, Inc.
Portsmouth, NH
