"Dr. Volker Jaenisch" <[EMAIL PROTECTED]> writes: > Dr. Volker Jaenisch schrieb: >> I keep you informed of my findings. > You've been right - the double-defined statements are due to the > long==int64_t identity. > > After application of the attached changes the code does compile and link. > These changes have been done in a formal way (eliminate explicit > ocurrence of int64_t) but not by understanding the code itself.
Looks great, thanks a lot for testing this. At first sight I would rather remove the "long" rather than the "int64_t" functions, so it's much simpler to be portable across a wide range of archs. > So they are not ment to be a valid patch. Can you please review if > these changes do the right thing and will cause no problems. Some minor suggestions: - the default diff format sucks, try something like: "diff -Nur" instead. - once it works, delete the commented-out lines before creating the patch, they just add noise. Thanks again for trying carob on 64bits. PS: is the makefile change required? _______________________________________________ Carob mailing list [email protected] https://forge.continuent.org/mailman/listinfo/carob
