|
From: Benedikt F. <b.f...@gm...> - 2025-01-11 11:15:31
|
Am 11.01.25 um 09:51 schrieb Philipp Klaus Krause: > Embarassingly, a typo fix in a regression test hadn't been comitted to > the next branch yet, and thus didn't get picked to trunk, and we now see > test failures in the farm, since a test was enabled for one target > (pdk14) that doesn't have enough memory to link it. > > I really should have rerun the full tests on trunk after picking, before > comitting to trunk. > > This is the typo fix in the next branch, that we need in trunk now: > > https://sourceforge.net/p/sdcc/code/15221/ > > Philipp > > P.S.: I'm still looking into the remaining parts of bug #3814, which > also gives us a time window for a possible fix of bug #3646. Since I take it that bug #3814 was a genuine regression, I'd say that if you can quickly come up with a fix for that, you can sneak the typo fix into trunk, as well, before creating the RC2 tag. In the long run, however, I believe that we should really consider switching to a semantic versioning based branch model with e.g. a 4.6.0-dev branch from which bug fixes can be cherry picked for stable 4.5.x releases. Benedikt |