On Fri, Jun 15, 2007 at 10:46:04AM +0200, Raphael Hertzog wrote: > On Fri, 15 Jun 2007, Steve Langasek wrote: > > > This should give the same behaviour indeed. The modules meta-package would > > > be broken when the linux-image-2.6 metapackages are upgraded > > > unsynchronized.
> > Hmm; I guess I meant > > Depends: linux-image-2.6-686 (>= 2.6.21), linux-image-2.6-686 (<< 2.6.22) > > but I realize now that breaks when the ABI changes within an upstream kernel > > revision. > Indeed, and the version of the metapackages doesn't include an explicit > reference to the ABI. IMO it should, then we could do: > Depends: linux-image-2.6-686 (>= 2.6.21-1), linux-image-2.6-686 (<< 2.6.21-2) This seems like a good idea to me. What does the rest of the kernel team think? > BTW, the version numbers in the changelog doesn't correspond to the > version numbers of the generated package. While this is allowed, its there > a compelling reason to do so in this case? > $ zless /usr/share/doc/linux-image-2.6-686/changelog.gz > linux-latest-2.6 (7) unstable; urgency=low > * Update to 2.6.21-1. > * Remove etch transition packages. > -- Bastian Blank <[EMAIL PROTECTED]> Tue, 29 May 2007 14:26:20 +0200 > [...] > $ dpkg -l linux-image-2.6-686 | grep ^ii > ii linux-image-2.6-6 2.6.21+7 Linux kernel 2.6 image on > PPro/Celeron/PII/PIII/P4 Dunno. I don't see a compelling reason not to, either? -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. [EMAIL PROTECTED] http://www.debian.org/