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

Re: Questions about the amd64-multiarch-3 proposal



On Fri, Apr 16, 2004 at 12:37:08AM +0200, Bill Allombert wrote:
> I have some comment on the amd64-multiarch-3 proposal
> <http://people.debian.org/~ajt/amd64-multiarch-3.txt>
> First I think this update go in the right direction, at least it does
> not look like a large kludge, but I need some clarification.

Well, it doesn't actually work, since it doesn't provide a feasible way
to upgrade from sarge to that scheme.

> I will assume, while it is not stated explicitly that amd64-multiarch-3
> follow  amd64-multiarch-2 w.r.t. the introduction
> /usr/$(gcc -dumpmachine)/{lib,include}
> Without this assumption the proposal make no sense.

Yup. I'm only interested in the package management details, which are
mostly independent from the actual file structure.

> >        lib<foo><n>-<arch>
> >               - .so file
> >                - libfooN-x, libfooM-x, libfooN-y must all be concurrently
> >                  installable
> If you ship the .so symlink here, how will you have libfooN-x and libfooM-x
> concurently installable ? 

That's the .so.1.2.3.4 file (ie, the library itself), not the development
symlink.

> >        lib<foo><n>-dev-<arch>

(This should probably be lib<foo>-dev-<arch> in most cases)

> 2) Should -dev-common depend on -dev-arch (even after the transitional 
> period?)

Probably. That'd make it:

	foo-dev-common
		Depends: foo-dev
	foo-dev-amd64
		Provides: foo-dev
		Depends: foo-dev-common

or so.

Cheers,
aj

-- 
Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
Don't assume I speak for anyone but myself. GPG signed mail preferred.

Protect Open Source in Australia from over-reaching changes to IP law
http://www.petitiononline.com/auftaip/ & http://www.linux.org.au/fta/



Reply to: