On Tue, May 30, 2006 at 02:50:16PM +0200, Wouter Verhelst wrote: > On Tue, May 30, 2006 at 12:05:26PM +0200, Andreas Barth wrote: > > Timeline > > ======== > > Now, let's please take a more detailed look at the time line: > > Thu 15 Jun 06: > > last chance to switch to gcc 4.1, python 2.4 > > review architectures one more time > > last chance to add new architectures > > RC bug count less than 300 > Since m68k pretty much depends on the gcc-4.1 transition to make it in > again, I would suggest that we (as in, the m68k port) make the switch to > GCC4.1 as the default already. This will allow us to verify that stuff > actually builds and works, and to catch up with building those that fail > with ICE in gcc-4.0 before that time. Since m68k is not a release > architecture right now, this should not cause any problems for any other > port if the GCC 4.1 transition does not happen, but it will help if it > does. > Thoughts, objections? Since it seems gcc-4.1 is the only way to get m68k back up to building a decent fraction of the archive, I think it's fair to switch to gcc-4.1/g++-4.1 as the default now on m68k, yes. From everything I hear, it at least isn't going to be worse than the status quo. I still wouldn't count gcc-4.1 build regressions in packages as release-critical until at least one other architecture had switched to it as default, even if m68k was otherwise ready to go as a release candidate, but that shouldn't stop you from doing porter NMUs anyway. BTW, can you tell me anything about the dip in http://buildd.debian.org/stats/graph2-quarter-big.png for m68k? Seems to be heading in the wrong direction again for being a release candidate. I see 12 buildds actively uploading packages for m68k, is this too few or is there some other problem? Cheers, -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. vorlon@debian.org http://www.debian.org/
Attachment:
signature.asc
Description: Digital signature