On Wed, Jun 14, 2000, H. Peter Anvin <hpa@transmeta.com> wrote:
> Johannes Erdfelt wrote:
> > >
> > > ... and until they do, they are mostly a curiousity. If EFI turns
> > > ubiquitous, then we can see if we can use it to clean up booting. For
> > > now, it's a non-issue.
> >
> > You need to actually push something before it becomes accepted.
> >
> > Thankfully, this does not have to be ubiquitous before starting to
> > support it.
>
> It does before there is any point to it.
Ok, I think your opinion is clear.
Also to clarify, I completely disagree with that opinion.
> > EFI is much more than something as trivial as serial console support.
> > In fact, it won't even help you with serial console support before it
> > loads, like in the BIOS setup or something similar. That's not what it
> > was really designed for. Check the docs.
> >
> > That being said, anyone can install EFI onto their system, regardless if
> > it came that way from the vendor. In fact, you want something now? Grab
> > the sample implementation, compile it and install. I haven't tried this
> > myself, but Intel says it's possible.
>
> The point is that noone that matters (PC vendors) care, so you're not
> going to see it.
I think you'd be surprised what the actual reality of the situation is.
I unfortunately cannot go into further detail, but I can tell you that
"noone" is incorrect.
> > Anyway, I still don't see the reason for sinking resources into the
> > legacy boot loaders (LILO, Syslinux, etc) while there is a perfectly
> > reasonable alternative which is here today.
>
> It's not here today in any kind of usable form, meaning I can walk up to
> J. Random PC and boot it using EFI. It's a paper spec.
I really wanted to kill this thread since it quickly turned pointless,
but what you wrote is very misleading. There are indeed implementations
around including the sample implementation that I referred to in my
previous email.
There are boot loaders as well, including the one we have developed for
ia64.
Not ready for primetime, but it is very far from just being a paper spec.
JE
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Jun 15 2000 - 21:00:33 EST