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

Re: Colour & filesystem translators



On Thu, Nov 30, 2000 at 04:30:45PM +0000, David Neary wrote:
> Marcus Brinkmann wrote:
> > 
> > Am Don, 30 Nov 2000 15:49:24 schrieb David Neary:
> > > I compiled a custom hurd post-install, and installed an entry for it in
> > > menu.lst. When I booted this new kernel & module (I also recompiled
> > > GNUmach), all of the translators which were there before were no longer
> > > active, and I had to re-set them.
> > 
> > Are you talking about active or passive translators?
> > Active translators always vanish when you reboot, as they are
> > only processes. Passive translators never go away by themselve.
> 
> I'm talking about passive translators - things like mapping /dev/hd0s2
> to /home/dave.

This would be a serious problem, but I can't see how this could happen.
Passive translators are like symlinks, they are written into the filesystem.
Please try to carefully reproduce the bug and make a concise report if
it persists.

> > It's exactly what you say. The CVS and Debian kernel differ.
> 
> Any idea if/when the relevant changes in the Debian kernel which allow
> for colour will be included in the CVS kernel?

No. For example, oskit-mach doesn't have color support IIRC, and the patch
we use currently doesn't apply there, so it doesn't make much sense.
This is not really an argument to not apply it. The real reason is that
I am hoping for the colortext translator, which would make my patch
obsolete.
 
> When I used "modules", I meant translators - I got this from GRUB -
> "module /boot/serverboot.gz".

Mmh, okay. Those are two different things, though. A multiboot module is
something that is run by the kernel after initialization, so that you
don't need to code os specific stuff into the microkernel.

> As I said, I'm just starting to get used
> to the terminology at the moment. By the way, your article on
> translators at hurddocs.sourceforge.net is excellent. It went a long way
> toward answering some of my sillier questions :)

Hopefully it helps you to provide a better diagnosis of the translators
vanish when using a different kernel/hurd problem :)

Have fun,
Marcus



Reply to: