On 1/18/09, Richard Guenther <richard.guent...@gmail.com> wrote: > On Sun, Jan 18, 2009 at 10:35 PM, H.J. Lu <hjl.to...@gmail.com> wrote: > > Hi, > > > > This 4.3 regression: > > > > http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38905 > > > > is caused by a backport, which also introduced a 4.3 > > regression: > > > > http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36765 > > > It uncovered a latent problem. Of course doing that still makes > it a regression. > > > > Since we didn't backport testcases to 4.3 branch, we > > didn't notice 4.3 regression when the packport was > > applied to 4.3 branch. If we will backport bug fixes > > to 4.3 branch, shouldn't we always backport testcases > > for all 4.4 regressions? > > > This would be probably a good idea, but it is extra work and > may cause extra testsuite FAILs between releases that are > not regressions (simply because only the testcase is new).
We can limit new testcases, which we backport to 4.3 branch, only to those 4.4 regressions against 4.3 branch. They should always pass on 4.3 branch by definition. -- H.J.
