2.6.26 kernel in hardy?

[Matthew Flaschen]

H.S. wrote:
> Matthew Flaschen wrote:
>> H.S. wrote:
>>> In other words, it is not uncommon for a kernel upgrade to break a
>>> particular hardware in a system. The solution is invariably to revert to
>>> an older one.
>> No, it's not.  That's a temporary, often flawed, work-around.  The
> 
> It is temporary, but flawed or not, that cannot be said in general
> terms. Here "flow" can only be defined based on the situation at hand
> (see below).

Why don't you read people's emails before replying.  I said "often
flawed", not always flowed.

>> solution is to fix the new kernel.  The kernel devs have a policy of
>> never "giving up" on hardware people are still using.
> 
> I understand that. And those fixes are reflected with the newer minor
> number kernel versions. But, *till* those fixes are available, and one
> wants to use the hardware in question, there is no other choice, is
> there now? In other words, one has to use the latest working kernel
> version. Again, a no-brainer.

The no-brain part is that what you call the latest working kernel (i.e.
a kernel that support the feature you want), most likely will not work
with Hardy.

> Keeping all these factors mind, I wish we had older version kernels (for
> exampe, 2.6.26 here, which was used in alphas) available for Intrepid
> for those poor souls for whom the newer kernel does not work properly.

They are available "for Intrepid", they just most likely won't work. How
long do you expect them to forward-port outdated kernels, exactly?  Do
you not realize this wastes time that could be used to fix the actual bug?

Matt Flaschen