Re: [Valgrind-users] Clues on cleaning uninitialised value

[Xiang Y. <xy...@pr...>]

> On Thu, 10 Apr 2003, Xiang Yan wrote:
>
> > > > ==32225== Conditional jump or move depends on uninitialised value(s)
> > > > ==32225==    at 0x86823C2: CMP_Compare (in /home/xiang/test/m27/srv)
> > > > ==32225==    by 0x8684660: CMP_ModularReduce (in
> > /home/xiang/test/m27/srv)
> > > > ==32225==    by 0x868DBA7: Alg_ComputeModQ_GHash (in
> > /home/xiang/test/m27/srv)
> > > > ==32225==    by 0x86686CA: A_X931RandomGenerateBytes (in
> > /home/xiang/test/m27/srv)
> > >
> > > For a start, recompile the code in question with -g if you can, so you
at
> > > least get line numbers in your error messages.
> > >
> >
> > I'm using -g, no line number. I'm suspecting it's inside oracle oci lib.
Is
> > it possible?
>
> Er... is what possible?  Getting line numbers -- depends if you can
> recompile /home/xiang/test/m27/srv, whatever that is...
>
Thanks.
No, it's in debuging mode, couldn't get the line numbers somehow.  All
errors(total 2500) come from one oracle oci function call, so I have to give
up cleaning them :(