[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#616621: nmu: for libsigsegv transition



On Sun, 2011-03-06 at 01:31 +0100, Christoph Egger wrote:
> libsigsegv in unstable changes soname for the library so all depending
> packages need to be rebuild. The depwait is needed as 2.9-1 misses the
> actual shared library by accident (#616576) so packages build against
> that version will link against the static library. That's also the
> reason clisp is needed as it was build against -1

Other than the amd64 maintainer upload, clisp hasn't actually managed to
build anywhere yet, so can't be binNMUed.  I've added the dep-wait for
it anyway, which means it will be retried once the updated libsigsegv is
available.

Does the amd64 package need binNMUing as well?

> I did a test-rebuild of asymptote to double-check it will be working
> which was successfull.
> 
> Hope I did get the syntax right.
> 
> nmu clisp_2.49-2 . ALL . -m "Build against newer libsigsegv"
> nmu gnu-smaltalk_3.2-1 . ALL . -m "Build against newer libsigsegv"
> nmu asymptote_2.02-2 . ALL . -m "Build against newer libsigsegv"
> nmu scheme2c_1993.3.15.2-10 . ALL . -m "Build against newer libsigsegv"

fwiw, you can do all of the above in one command:

nmu clisp gnu-smalltalk asymptote scheme2c . ALL . ...

gnu-smalltalk has been marked as Failed on powerpc, s390 and sparc for a
few months due to #590227 (although previous versions built fine) so
hasn't been binNMUed on those architectures.  Is there any reason to
suppose that it will now build?

> dw clisp_2.49-2 . ALL . -m 'libsigsegv (>= 2.9-2~)'
> dw gnu-smaltalk_3.2-1 . ALL . -m 'libsigsegv (>= 2.9-2~)'
> dw asymptote_2.02-2 . ALL . -m 'libsigsegv (>= 2.9-2~)'
> dw scheme2c_1993.3.15.2-10 . ALL . -m 'libsigsegv (>= 2.9-2~)'

dep-waits need to be on a binary package; I chose libsigsegv2 in this
case.

Regards,

Adam




Reply to: