Re: Status of BUILD_REAL_64_BIT_MODE?
"ZIGLIO, Frediano, VF-IT" <[email protected]>
| Newsgroups | gmane.comp.db.unixodbc.devel |
|---|---|
| Message-ID | <[email protected]> |
> > I'm wondering how sane it is to consider turning on > BUILD_REAL_64_BIT_MODE in a 64-bit Linux distribution (I'm responsible > for Red Hat's packaging of unixODBC, if you're wondering). > The comments > in sqltypes.h about it are pretty discouraging, but maybe > that's obsolete > information now? > > I trolled the archives a bit and found > http://mailman.unixodbc.org/pipermail/unixodbc-dev/2008-Januar > y/001139.html > which suggests that Debian is enabling it but found several additional > fixes necessary. Comments on the correctness/completeness of that > patch would be especially useful. > > regards, tom lane I wrote the mail :) IMHO we should push BUILD_REAL_64_BIT_MODE like Debian does. The main reason is that main ODBC platform (win64) use it! Converging all to a single definition is better. After many posts to Microsoft I get this page http://msdn.microsoft.com/en-us/library/ms716287(VS.85).aspx! There are currently some issues (for instance SQLROWSETSIZE should be SQLULEN, not SQLUINTEGER) but mostly should work. The real problem is drivers which should be checked better but I think that if Debian is compiling unixODBC with BUILD_REAL_64_BIT_MODE defined they already fixed many issues for you :) Frediano Ziglio _______________________________________________ unixODBC-dev mailing list [email protected] http://mailman.unixodbc.org/mailman/listinfo/unixodbc-dev
smime.p7s
(application/x-pkcs7-signature, 3.1 KB) - not displayed