Re: I2O detection problem for 2.6 kernels
On Friday 25 November 2005 21:55, Frans Pop wrote:
> On Friday 25 November 2005 22:07, Lennart Sorensen wrote:
> > i386 has dpt_i2o, while amd64 being 64bit does not. Does discover know
> > that? The driver apparently isn't 64bit clean and is flagged in the
> > kernel config as such. So while dpt_i2o is preferable, it is not an
> > option on amd64.
>
> Thanks for that info.
>
> Looks like discover does have an option for architecture specific
> overrides, but there are currently none for amd64. Not sure though as I'm
> not really familiar with discover internals.
>
> If this override is set, it will mean that amd64 porters will need to keep
> and eye out and request for it to be removed should the driver get fixed
> in the kernel.
>
> > This does seem like an installer problem.
>
> However, I'm not sure if it's worth doing this as the i2o_block does not
> seem to function properly given the problems with partitioning mentioned
> a bit later in David's mail (which is what I was really refering to with
> my comment).
> So IMO there still is a kernel problem that prevents installation...
>
> Cheers,
> FJP
>
> P.S. In order to be able to properly track this issue an installation
> report [1] should be filed with all the relevant info from the mails so
> far. David: could you do that please?
> Please also try to get the lspci info.
Doing a reportbug is not possible on the machine as I can not get it
to install. And using reportbug from this laptop is no good as the list
of packages would be entirely misleading (its i32).
I would happily use lspci IFF it were available on the installer system.
But it is not. It does strike me that adding this would be a small price
to pay in space and gain enormous benefits.
David
>
> [1] http://d-i.alioth.debian.org/manual/en.i386/ch05s03.html#submit-bug
Reply to: