Re: pkgsrc NetBSD 6.0/x86_64 2013-06-11 00:30

"Mike M. Volokhov" <[email protected]> Sun, 16 Jun 2013 04:59:12 +0300
Newsgroups gmane.os.netbsd.devel.pkgsrc.wip.general
Message-ID <[email protected]>
16 June 2013, 00:32, Aleksej Saushev wrote:

> "Mike M. Volokhov" <[email protected]> writes:
> 
>> 15 Jun 2013, 22:03, Aleksej Saushev wrote:
>> 
>>> Krister Walfridsson <[email protected]> writes:
>>> 
>>>> On Sat, 15 Jun 2013, Aleksej Saushev wrote:
>>>> 
>>>>> I doubt that posting build results for development packages on legacy platform
>>>>> is useful to anyone. Is it possible to convert to 6.1 at least?
>>>> 
>>>> I am planning to upgrade to 6.1 soon.
>>>> 
>>>> But it is wrong to call 6.0 a "legacy platform" -- 6.0.2 is just
>>>> released, and even 5.2 is still a supported platform for pkgsrc.
>>> 
>>> Whatever you call it (even 6.0.1 release has made it obsolete),
>>> 6.0 release is plain broken for work related to number crunching
>>> whether we like it or not.
>> 
>> From numbering perspective it's better to keep packages for elder
>> platform rather decent system, simply because of backward compatibility.
>> However, for bulk builds of *the* pkgsrc it should be irrelevant -
>> the pkgsrc poses itself as really multiplatfrom packaging framework,
>> isn't it?
> 
> No, it isn't irrelevant. NetBSD 6.0 release contains critical bug that
> precludes building of relevant software. Working around a bug just
> to support _now_obsolete_ version of OS is stupid. Packaging framework
> has nothing to do with it, it is the software that doesn't get built.
> If we want to know how packaging framework behaves, we should run
> regression tests instead.

I'm sorry, perhaps I missed point here. What is the critical bug you're
constantly referring to?

--
Mishka


------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev