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

Bug#299074: Fw: Re: SunBlade D-I problems

Maybe someone could have included me on this bug discussion. The date of
my filing the bug, I downloaded the install-sparc/current/ image to use.
I'm not sure what version that is.

I believe it is rc3, but it may be rc2 (when was rc3 released?). I do know
that I used 2.6 images, and maybe that it is the issue here. Is it
possible that the detection mechanism does not work with 2.6?

I'll agree that autoloading the modules isn't the best thing, but unless
the detection can be fixed, it may be the only solution.

Even something like:

if prtconf -v | grep -q SUNW,esp;
	modprobe esp

Would suffice. (note, that's a simple solution, there are other names that
esp comes under).

It's also possible that the above note is the problem. The 4500's use
SUNW,fas as the prom name of the esp scsi controller, but that wouldn't
explain the hme ethernet loading.

On Mon, Mar 14, 2005 at 07:22:50PM -0800, David S. Miller wrote:
> Can you let them know what image you were using in
> bug #299074?
> It's important that the auto-detection of SBUS devices
> works, just autoloading them is just a bandaid of course ;)
> Thanks Ben.
> Begin forwarded message:
> Date: Mon, 14 Mar 2005 22:21:53 -0500
> From: Joey Hess <joeyh@debian.org>
> To: Jurij Smakov <jurij@wooyd.org>
> Cc: debian-sparc@lists.debian.org, 299074@bugs.debian.org
> Subject: Re: SunBlade D-I problems
> Jurij Smakov wrote:
> > >>[3] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=299074
> > >
> > >He lists the exact way to solve the problem, which is that
> > >SBUS driver modules don't get loaded properly, and that
> > >by autoloading the modules he listed the problem can be worked
> > >around.
> > >
> > >It looks to me like there is an autoprobing and automatic
> > >module loading mechanism for PCI devices, and there is not
> > >one for SBUS devices.
> > 
> > Autoprobing for SBUS devices has been included in discover1 for quite some 
> > time now. And it works nicely with the recent (pre-RC3) installer images, 
> > I have tested it myself just a few days ago on an Ultra 1. Unfortunately, 
> > the submitter does not mention the version of the installer he used...
> > Anyway, as I was writing this message, the CIA bot reported that Joey Hess 
> > has made a commit forcing the loading of sunbmac, sunhme and esp modules 
> > to close this bug :-).
> Yes, but I didn't realize that it was probably a non-bug. Do you think I
> ought to revert that then?
> I guess the thing that's worrying me is that the last recorded discover1
> change for sbus was this:
> discover1 (1.6.6) unstable; urgency=high
>   * Joshua Kwan
>     - Fix broken sbus detection on sparc64.
>  -- Joshua Kwan <joshk@triplehelix.org>  Thu, 26 Aug 2004 22:25:31 -0700
> Ben didn't mention in #299074 what version of d-i he used, but it seems
> unlikely that it predated rc2, which should have included that change.
> -- 
> see shy jo

Debian     - http://www.debian.org/
Linux 1394 - http://www.linux1394.org/
Subversion - http://subversion.tigris.org/
WatchGuard - http://www.watchguard.com/

Reply to: