Bari Ari <[EMAIL PROTECTED]> writes:

> Eric W. Biederman wrote:
> 
> > Ronald G Minnich <[EMAIL PROTECTED]> writes:
> > 
> >> to boot windoze: best bet I think is to boot winsucks with two-kernel
> >> monte or similar tool. I.e. use Linux to boot winsucks, not linuxbios. 
> >> 
> >> If you want to do that for us ... well, nobody would complain :-)
> > 
> > 
> > If anyone takes this up I would suggest booting and running DOS first.
> > Dosemu and a few other open source projects have this ability so what
> > is required is not a secret.  Once you can run DOS windows should just
> > be just a few steps away...  Of course windows might be easier...
> > 
> I'm not surprised that this has come up again since LinuxBIOS was on /. 
> again this past weekend. Maybe there's another project here very 
> seperate from LinuxBIOS to get an open BIOS to boot and fully support M$ 
> systems? Seems there are lots of systems out there that could benefit 
> from an open source BIOS that supports the legacy of windoze. But I 
> already feel sick just thinking about it.

:)

The only nice thing is that BIOS's have enough different implementations
that you don't need to be bug compatible, which reduces the amount of work
needed by several orders of magnitude.

For the most part for linux 2.2.x we need to do enable all of the
hardware devices.   And even 2.4.x doesn't do enable all of the
hardware it should.  So I do see some advantage in having a
mkelfImage-bkcompat that builds and Elf image that has the backwards 
compatiblity code in it.

I don't see any real advantage in actually booting windows,
but I do see some advantage in having the infrastructure in place to 
have the option of booting windows, or dos.  Being able to do a little
more generally means you have a good solution.

The truth is Windows looks like it enables the hardware better then
even linux-2.4.x.  But I'm not going there.

Eric






Reply via email to