On Thu, 6 Jan 2011, Ian Lance Taylor wrote: > > 5. Similarly, if I mark a bug as known to work in 4.5.2, will this > > lead to it eventually being closed? > > Yes, when the 4.4 release is closed, bugs known to work in 4.5 will be > closed.
That's not how it works. I did the last couple of branch closures (4.1 and 4.2); I only went through the regression bugs or open bugs with a milestone on the closed branch, and for those bugs interpreted the bug log to determine whether they should be retargetted or closed. For any other bug (i.e. any non-regression), if it works on 4.5 and trunk it should be closed, and if it works on 4.5 but not on trunk it should be marked as a regression. So marking a bug as known to work in 4.5 is only useful if you also mark it as a regression in 4.6 (which of course means testing on trunk) (or, as the case may be, as a 4.3/4.4 regression from some older release that was fixed in 4.5 but may yet be fixed in 4.3/4.4). If it works in 4.5 and is not a regression on trunk or an open branch, it should just be closed; we don't keep bugs open for things that work on trunk unless they are regressions in a release branch. There are very many UNCONFIRMED bugs. If you can confirm that such a bug really is a bug in GCC and is still present on trunk, then moving it to NEW is useful. -- Joseph S. Myers jos...@codesourcery.com
