On Mon, 2005-11-21 at 22:20 +0100, Markus Trenkle wrote:
> Lincoln,
>
> Am 16.11.2005 um 05:04 schrieb Lincoln A. Baxter:
>
> > If you are willing to test a new Makefile.PL after I update it,
>
> I will do so. My modifications to the Makefile are simple, I only
> removed the compiler options in the linker flags. For some reason the
> newly generated Makefile is completely different, so I have to
> describe what I did. The original options are still in the comments
> at the beginning of the file. There it reads
>
> # dynamic_lib => { OTHERLDFLAGS=>q[ -Wl,+s -Wl,+n +DA2.0W +DS2.0
> -L/sorry/oracle/app/oracle/product/10.1.0/rdbms/lib/ -L/sorry/oracle/
> app/oracle/product/10.1.0/lib/ -lclntsh `cat /sorry/oracle/app/
> oracle/product/10.1.0/lib/ldflags` `cat /sorry/oracle/app/oracle/
> product/10.1.0/lib/sysliblist` -lm ] }
>
> I changed the line further down to
>
> OTHERLDFLAGS = +s +n -L/sorry/oracle/app/oracle/product/10.1.0/rdbms/
> lib/ -L/sorry/oracle/app/oracle/product/10.1.0/lib/ -lclntsh `cat /
> sorry/oracle/app/oracle/product/10.1.0/lib/ldflags` `cat /sorry/
> oracle/app/oracle/product/10.1.0/lib/sysliblist` -lm
>
> If this is not enough, I will reproduce what I did in the first
> place. But I dont have access to the server now.
>
> Markus
>
Ok, so it looks like the Makefile.PL did not recognize that you were
using the gcc (linker) instead of ld or the linker the HP C compilers
(like the soft bench compiler) invoke.
Question: If you no long have access to the server how would you test
anything I produce? By the way, for what it's worth, I no longer have
access to any HP machines. So, while I still know some things about
them, with out the ability to test, I can not do much serious
development. As it is I would only be able to make minor changes,
which OTHERS would have to test.
Let me know how you want to proceed.
Lincoln
