Proposal for testing branch

Nathan Coulson conathan at conet.dyndns.org
Thu May 20 12:25:41 PDT 2004


>
> "Matthew Burgess" <matthew at linuxfromscratch.org> wrote in message
> news:20040520194547.1b38f65d.matthew at linuxfromscratch.org...
>> On Thu, 20 May 2004 11:54:58 -0500
>> "Larry Lawrence" <larry at linuxfromscratch.org> wrote:
>>
>> > I don't see any cons, but I'm sure they can be added to this thread
>> > for discussion.
>>
>> Larry, please see my summary of the "Building Kernel" thread at
>> http://www.linuxfromscratch.org/pipermail/lfs-dev/2004-May/045945.html.
>> NPTL *will* be in 6.0
>>
>> Cheers,
>>
>> Matt.
>
> Please read what I said carefully -
>
> "Second, it would allow a reasonable 6.0 branch to be developed if a
> release
> is needed prior to implementing the new threading "
>
> If you commit to NPTL prior to it being cross platform, I think your
> making
> a mistake, however I am by no means demanding it to be so.  I'm proposing
> that it does not go in testing right away. For what purpose is testing if
> you don't introduce one major change at a time and test it.  By the time
> 6.0
> comes up, everything could be a go.  If there are delays upstream, why
> trap
> yourself into waiting for NPTL before you ever release 6.0 (or scrap the
> cross platform support).
>
> Larry

Just to interrupt, the bootscripts have included udev in the main
bootscripts, assuming [sorry] that it would be included in the next book
version.

(Easy enough to move back I suppose, but more food for thought).



More information about the lfs-dev mailing list