On Sun, 2002-06-23 at 16:55, Felipe Alfaro Solana wrote:
> On Mon, 2002-06-24 at 01:30, Jeffrey Stedfast wrote:
> > Given that:
> >
> > 1. Evolution works fine on kernel 2.2
> > 2. Evolution works fine on kernel 2.4
> > 3. Evolution does not work fine on kernel 2.5...
> >
> > Seems pretty obvious to me where the problem is - kernel 2.5, it is the
> > only thing that has changed in the equation.
> >
> > well, unless kernels 2.2 and 2.4 were broken in such a way that made
> > Evolution work :-)
> > But assuming that was the case, then it could be argued that kernel 2.5
> > *is* broken for changing behavior.
> >
> >
> > Anyways, kernel 2.5 is a development kernel, so you are supposed to
> > expect things to not work. We should not have to make sure that
> > Evolution works on development kernels.
> >
> >
> > You are of course free to debug this, I would probably start with
> > debugging ORBit since if it really is not a kernel bug, then the bug is
> > mostly likely in ORBit.
> >
> > Jeff
>
> Hello, Jeff...
>
> What scares me all about this is the fact that the kernel people still
> go on patching and developing the 2.5 kernel, but the problem is still
> there. It seems like if no one cares... I wonder what might have changed
> and I would like to help in finding the culprit but to be honest I think
> I don't know how... (I'm not neither a kernel hacker nor a evolution
> hacker).
Hi Filipe,
I have already responded to this question on LKML, but I'll
add my comments here.
Mark McLoughlin and Michael Meeks agree that this bug will need
to be tracked down in ORBit 1.x. It is possible that the
problem will go away when Evolution gets ported to Gnome 2.0,
since there are lots of changes in ORBit2. Neither Mark or
Michael have development machines running a 2.5 kernel, so it
is difficult for them to reproduce this bug. Also, everyone
who does Gnome work is busy with the Gnome 2.0 release effort.
It is possible that once Gnome 2.0 is released, they may find
time to build and install a 2.5 kernel for testing. On the
other hand, they may only get around to testing this closer
to the arrival of the 2.6 stable kernel. I am guessing we
won't have a 2.6 kernel for at least another six months.
I have tried to track down this bug, but ORBit debugging is
quite a black art. It would probably take much more
time for the ORBit dudes to help us track the bug down
than it would for them to do the work themselves.
As to whether this is an ORBit or kernel 2.5 bug, it could go
either way. It may be that ORBit is relying on older kernel
behavior that should not be used. In this case, we'd need an
ORBit fix. On the other hand, it could always be kernel 2.5
bogosity, too.
So, I think we just need to be patient.
Miles
_______________________________________________
evolution-hackers maillist - [EMAIL PROTECTED]
http://lists.ximian.com/mailman/listinfo/evolution-hackers
