On Tue, May 01, 2007 at 01:29:01PM +0100, Claus Reinke wrote: > > sometimes i feel like a human buildbot.. > > would it be possible to set some maximum failure time (1 week or so), > so that buildbot failures for longer than that would cause all new patches > to be put on hold until the head+libs _build_ again on all major platforms?
FWIW, that wouldn't have helped the build system fixes get done any quicker, and the other problems tend to be short-lived. Incidentally, the HEAD has mostly been buildable; it's only if you put extra, unbuildable libraries in the tree that it's not been working, for which there is a simple workaround! > it is an unusal time with all those build system changes, but similar > effects > have been caused by other changes in the past, for windows builds at least. We're hoping that now that buildbot gives us regular Windows builds that (a) Simon can access the build trees of if there's a problem, and (b) that run the testsuite, Windows-specific problems will be noticed and fixed quicker. Similarly we now have OS X nightly builds running the testsuite, so problems there should be caught earlier too. > if one could rely on at least one 'all buildable' window every week, > that would be nice. I don't know how you'd do that, short of reverting all patches since the last window if it doesn't build at the start of the next. You could imagine some sort of two-level system, where patches are only pushed to a guaranteed-buildable repo if a nightly build with them included goes through. However, I hope that in general such a system won't be needed, and it would be some effort to set up and maintain. Thanks Ian _______________________________________________ Cvs-ghc mailing list [email protected] http://www.haskell.org/mailman/listinfo/cvs-ghc
