Re: Help for Pro*C problem on Apollo (ACK)

In article <1992Oct27.175548.3913_at_ncube.com>, sbh_at_ncube.com (Shawn Honess) writes:
> We have written some tools in Pro*C on an older Apollo system (ack).
> I am trying to make them work on a newer Apollo, but I am getting some
> very strange problems.
> [...]
> When run normally, I get a "reference to illegal address" error
> from the OS, and a crash. Running the program in a debugging
> environment reveals that the crash occurs in the "sqlsca(&sqlca)"
> call; the first line of actual ORACLE code!!! The error message
> which the debugger gives is:
>
>> Segmentation fault in sqlalc.sqlhch at \

      line 358 in file "//phoenix/src1/6030/sqllib/sql/sqlalc.c"

>> couldn't read "//phoenix/src1/6030/sqllib/sql/sqlalc.c"


> 

>     This is the strange part.  We don't have, and never have had a

//phoenix must be the name of the machine at Oracle which had their source code. The debugger is hoping the machine's network is connected to yours, because it'd like to display the source file.

> I am able to compile and link successfully although I get the
> following warnings:
>
>> server - sgalen: P:4096,3,6
>> oracle: invalid sgasize argument: P:4096,3,6

What gives you these warnings?

>> No errors, no warnings, 143 info messages, C compiler 68K Rev6.7(316)
>> All globals are resolved.
>> There are unallocated .bss globals.

This looks like you're using /com/cc -- try compiling with /bin/cc.

> I have spent plenty of time RTFM and tried to call in favors
> from those I know at ORACLE, but I'm still stumped. I'd appreciate
> anyone who could point me in the right direction. My guess is that
> I have some kind of version conflict.

Probably. I think Oracle used /bin/cc to compile their libraries, with pretty boring options. /com/cc usually defaults to some things that can't be done on "standard" unix, and there can be conflicts when mixing them. Unfortunately, all the pleading to Oracle is useless, because then have abandoned the Apollo platform.

If merely using /bin/cc doesn't help (or you can't figure out the conversion), e-mail me. We do lots of Pro*C on Apollos..

-- 
Frederick G. M. Roeber | CERN -- European Center for Nuclear Research
e-mail: roeber_at_cern.ch or roeber_at_caltech.edu | work: +41 22 767 31 80
r-mail: CERN/PPE, 1211 Geneva 23, Switzerland | home: +33 50 42 19 44
--  
"Sorry, baby, I can't take you to the pizza joint tonight, I've got to go
back to the lab and split the atom." -- Ayn Rand, "What is Romanticism?"