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

Re: Really 2.6.18?

martin f krafft <madduck@debian.org> writes:

> Are we sure we want 2.6.18 as the kernel for etch? I reported two
> bugs, #391929 and #391955, the first of which is readily
> reproducible on 2.6.18 only (including ABI -2), meaning I cannot see
> the problem with 2.6.17. #391955 is rather sporadic.
> I know the kernel team has been incredibly busy, but I have received
> zero reaction to my bug reports, which makes me think that they may
> not have been seen? After all, I did originally assign them to the
> kernel packages causing the problems: linux-image-2.6.18-1-amd64 and
> linux-image-2.6.18-1-686, rather than the linux-2.6 source package;
> they're reassigned now.

At least for XEN 2.6.17 has serious problems. I have three machines
that are unable to boot with 2.6.17 and them work fine with 2.6.18.

> Do we really want to release 2.6.18 with etch? If I alone am already
> able to identify two hard kernel freezes, there must be plenty
> others, no? Do we want to lock out users into 2.6.18 with its bugs?

Well, we'll probably need to rely on bugfix releases of 2.6.18 after
etch is release too.

> Wouldn't it be better to let 2.6.18 mature a bit more, provide
> 2.6.17 with etch, and let 2.6.18 follow with r1?

I think it's worse. Doing that, we'll probably have worse regressions
like machines that worked with 2.6.17 and stoped to be able to install
with 2.6.18.

        O T A V I O    S A L V A D O R
 E-mail: otavio@debian.org      UIN: 5906116
 GNU/Linux User: 239058     GPG ID: 49A5F855
 Home Page: http://www.freedom.ind.br/otavio
"Microsoft gives you Windows ... Linux gives
 you the whole house."

Reply to: