Re: vcs-tree 0.3.1-1 to etch, please
Kevin,
On Thu, Dec 28, 2006 at 09:59:29PM -0700, Kevin Rosenberg wrote:
> My package, vcs-tree 0.3.1-1, has a significant improvement compared to the
> etch version of 0.2.2-4. It uses a new form of delivery that won't
> break when SBCL changes binary "fasl" formats.
> However, 0.3.1-1, is stuck in unstable because of an SBCL bug on the
> sparc architecture which prevents this improved version from building.
So, what is being done about this bug? We're not going to ship vcs-tree
0.3.1 on all other archs, and ship vcs-tree 0.2.2 on sparc. Either the sbcl
bug needs to get fixed, or the vcs-tree sparc binary needs to be removed
from unstable with the agreement of the sparc porters.
Does the bug in question make sbcl unusable in general on sparc, or just
prevent building vcs-tree? Does the bug also prevent building the old
version of vcs-tree? (Is this the same bug preventing the current version
of sbcl from being built on sparc?)
> However, I think etch should include 0.3.1-1 rather than
> 0.2.1-3. Besides being robust against updates to the installed SBCL
> runtime library, 0.2.2-4 doesn't actually work in etch due to the
> incompatibility of SBCL's runtime. On etch the following error occurs
> with vcs-tree 0.2.2-4 and the sbcl version in etch:
> $ vcs-tree
> [...]
> #<SB-SYS:FD-STREAM for "file /usr/bin/vcs-tree" {A859FB9}> is in
> native code
> fasl file format version 55, but this version of SBCL uses format
> version 69.
AFAICS, this means vcs-tree has an unfiled serious/grave bug in etch
(serious: wrong dependencies; grave: unusable package). Please file this
bug in the BTS so we have a record of the problem while trying to sort out
what to do with vcs-tree 0.3.1-1.
--
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/
Reply to: