Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Mailing Lists -> Oracle-L -> PRO*C compiling AIX 5.3
In July, Brian Shanblatt posted a query to this list regarding a problem
with 9.2 Pro*C and AIX 5.3 for which there was no adequate answer
(http://dba.5341.com/msg/81329.html).
However, his post turned up in a Google search when I ran into the same error with the same components and this helped me understand that I had a soluble problem which I have now overcome. I wanted to add some value by posting a follow up.
Brian, I recently ran into this same issue with a properly installed version of Oracle 'Pro*C/C++: Release 9.2.0.1.0' on AIX 5300-06. It appears that between AIX 5300-04 and 5300-06 IBM put a section into /usr/include/standards.h with this #warning line and Pro*C/C++ doesn't like it. This warning then appears to throw off the entire precompilation phase such that simple types such as size_t aren't defined and everything falls in a heap.
After unsuccessfully trying various fiddles involving DEFINE= in my proc invocation, I decided that rather than modify standards.h in /usr/include, I would make a local copy of standards.h in the directory with my source code and header files and edit that.
I changed IBM's:
#if defined(__IBM_PP_WARNING)
#warning The -qdfp option is required to process DFP code in headers.
#else
#error The -qdfp option is required to process DFP code in headers.
#endif
into:
#if !defined(__IBM_PP_WARNING)
#error The -qdfp option is required to process DFP code in headers.
#endif
My proc invocation already included INCLUDE=., so this local version of standards.h got pulled in instead of IBM's one and the proc command in the makefile then went through successfully.
HTH,
KEN WALLIS
Technical Consultant
PRO*C compiling AIX 5.3
2007-07-19 - By Brian Shanblatt
Hi Peter and all.
Thank you for your reply.
I just did an install of the 9.2.0.8 patchset, which upgraded and
relinked
everything.
The same problem occurred with pro*c on the standards.h file.
I then relinked the precompilers and the shared client libraries.
Same error.
It might have something to do with the AIX patchsets. Metalink lists a whole bunch, and I will have the sysadmins on the machine check that we have them.
Comments?
-- http://www.freelists.org/webpage/oracle-lReceived on Mon Nov 26 2007 - 19:09:28 CST
![]() |
![]() |