On 22-06-2006 13:02:54 +0200, Stefan Manegold wrote:
There's one little show stopper for me, which is M5/SQL, which is not so green at all. I don't know where the fix will be. If it will be in M5 code, nothing has to change, but if the fix will be in SQL code, then I'd like to have that fix included in the release. It all depends on how quick this bug can be found, of course. I guess some assessment needs to be made with regard to this in the light of the M5 alpha release.
I was purely talking about the MonetDB 4.12, MonetDB/SQL 2.12, MonetDB/XQuery 0.12 release, i.e., M4 release branches, only.
M5 and hence M5/SQL are based on the development trunks (CVS HEADs) of MonetDB[4], sql and monet5, and hence are themselves as well as the respective M5 alpha tarball-only release independent of the M4-based MonetDB 4.12, MonetDB/SQL 2.12, MonetDB/XQuery 0.12 release.
IMHO M5/alpha should run against stable, not current. It makes no sense to rerelease the whole bunch of files, especially not if you suppose to do it at the same time. I've been running/compiling/testing M5 against stable for the last week(s).
Hence, M5/sql propoblen are ("luckily" ;-)) no showstopper for the M4-based release.
Appart from that, the M5/sql problems seems to be "only" a compilation problem of sql/src/backends/monet5/ with icc:
You mean... "only" = some platforms are completely red, and not a single Java based test succeeds. Anyway, let's kill the flames, you're the release engineer. I'll patch/fix afterwards again.