Re: 4.0.5: Merge Request

> >An easily reproducable segfault in a common PHP extension is a serious > >issue which could lead to potential security breaches and thus lots of bad > >mojo from nasty bugtraq postings. If we know about such a segfault and we > >have a fix and go ahead and release a "stable" package without this fix > >then our entire QA process is a joke. If we find such bugs before a > >release, we fix them. That's what the process is for. > > Have you checked the bug database lately? > There are 43 open reports with bug type 'Reproducible crash'. > There are actually even more of them. I can easily reproduce > at least 10 bugs which cause segfaults. Those haven't been fixed yet. > Those haven't been fixed for a couple of releases now. > So, with your logic, we shouldn't release before all of them are fixed? You ignored what I said, "If we know about such a segfault and we have a fix" And yes, if you can actually easily reproduce a segfault in a mainstream module right now, then we have a showstopper and we hold until it is fixed. Please provide specifics. Most of the ones I have looked at are platform and library version specific and either not easily reproducible nor common. > The QA process as it is IS a joke. Without the support from the developers > there aren't any possible ways that it can ever succeed. > It isn't the QA people who fix bugs. They just test and report to developers > who should FIX those bugs. Some core developers seem to have forget this.. QA is there to assure quality, not push out releases at all costs. Pushing out a release without this imap fix would produce lower quality rather than higher. That should be the first priority of QA. > What I would like to see now is a code freeze so that people would have > to start fixing bugs instead of creating new features. Otherwise this > really is a neverending circle. But this will never happen, I guess. > Could we vote for this? :) You are getting way off topic here. You objected to a segfault fix being merged into the release branch and thus slowing down the process. This was not a new feature. For the most part no new features have gone into the release branch. > I know it's more fun to create something new. But at least for > me it's a matter of honour that my code works as it's intented to..you > seem to disagree. Not at all. -Rasmus