On Wed, 1 Mar 2006, H.Merijn Brand wrote:
h.m.brOn Wed, 1 Mar 2006 12:09:27 -0500 (EST), "John L. Allen" <all...@grumman.com>
h.m.br> On Wed, 1 Mar 2006, Dave Mitchell wrote:
h.m.br> davemOn Wed, Mar 01, 2006 at 09:59:19AM -0500, John L. Allen wrote:
h.m.br> davem> We are using 32 bit Oracle 126.96.36.199 client and server, but I don't see
h.m.br> davem> why it shouldn't be able to be made to work with a 64 bit build of
h.m.brHere is where I read, and I assume Dave too, a 64bitall perl.
h.m.br64bitint, if supported by the OS, doesn't make it a 64bit perl.
h.m.br> davem> perl.
h.m.br> davemI would be amazed if it was possible to use a 32-bit oracle client
h.m.br> library davemwith a 64-bit perl executable.
h.m.br> davemI strongly suspect that's where your problem lies.
h.m.br> Normally I'd agree, but my 64 bit perl is sort of a hybrid: all the perl
h.m.br> binaries and extensions are 32 bit, as is the AIX 5.1 kernel. It's just
h.m.br> that perl was built to use 64 bit integers and 128 bit floats (and the
h.m.br> 128 bit C library), via the -Dusemorebits configure flag. This differs
h.m.br> from the -V setup H.Merijn just sent because he is using -Duse64bitall,
h.m.br> which gives him 64 bit binaries as well. Or, such is my understanding.
h.m.brWith Oracle, I would play safe (or it will play you).
h.m.brEither go for vanilla 32bit or for full 64bit.
Ultimately, you may be right. I know it works fine with 32 bit perl in my
case, so that is a way out, but I was really looking forward to using
64/128 bit ints/floats in perl... Perhaps a static build of DBD::Oracle
will help, and I will play around some more with various compile and link