diskless boot roadmap (was:L Re: GENERIC build broken )

1999-11-02 Thread Mike Smith

> On 1999-Nov-03 15:59:38 +1100, Matthew Dillon wrote:
> >:Well, bootp in the kernel has to die too.
> >
> >Huh?  And replace it with what?  BOOTP is the only way to get an NFS
> >root and swap.
> 
> Sun uses reverse ARP to do this.  Reverse ARP _is_ a hack, but it _is_
> an alternative to BOOTP.

RARP would be a really bad move at the moment, since everyone else is
using DHCP.  For those that are still following, the roadmap looks like
this at the moment:

 - On the PC, we will be following the PXE lead, since that's where the 
   market is going.  This basically involves using the PXE firmware 
   which will talk to a PXE server to get the loader.  It can also be 
   conned into using normal DHCP.  File transfer is either by MTFTP 
   (multicast TFTP) or normal TFTP.  The loader will do plain DHCP and
   use NFS to get the kernel straight out of the prospective root 
   filesystem.

   Please don't complain about how ugly PXE is.  It's the closest thing
   the PC world has ever seen to a sane network boot architecture.  
   You're welcome to go work on RPL if you want some perspective. 8)

 - On the Alpha, diskless booting involves BOOTP/TFTP to fetch the 
   loader.  The loader will then happily use BOOTP or DHCP again (since 
   SRM is too stupid to tell _us_ what it discovered) and NFS to read
   the kernel from the root filesystem.

As you can see from the above, there's no call for RARP, bootparam, RPL 
or any other form of mumbo-jumbo.  There's also no call for the kernel 
to do any of this work; the loader will do it all and pass it in to the 
kernel in a format it can use directly.

What is currently missing, you ask?

 - The loader DHCP code needs to be written.
 - The kernel needs to improve the way it handles interface 
   configuration information passed to it by the loader.
 - We need more PXE work done for the i386 loader.  Intel are helping 
   here a little, but I could use a couple of extra hands to get this 
   done sooner rather than later.

That's about it; we're almost there.

-- 
\\ Give a man a fish, and you feed him for a day. \\  Mike Smith
\\ Tell him he should learn how to fish himself,  \\  [EMAIL PROTECTED]
\\ and he'll hate you for a lifetime. \\  [EMAIL PROTECTED]




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: diskless boot roadmap (was:L Re: GENERIC build broken )

1999-11-02 Thread Peter Jeremy

On 1999-Nov-03 16:15:03 +1100, Mike Smith wrote:
>> On 1999-Nov-03 15:59:38 +1100, Matthew Dillon wrote:
>> >:Well, bootp in the kernel has to die too.
>> >
>> >Huh?  And replace it with what?  BOOTP is the only way to get an NFS
>> >root and swap.
>> 
>> Sun uses reverse ARP to do this.  Reverse ARP _is_ a hack, but it _is_
>> an alternative to BOOTP.
>
>RARP would be a really bad move at the moment, since everyone else is
>using DHCP.

I'm not suggesting we move to RARP.  Matt was stating that BOOTP is
the only way.

>  There's also no call for the kernel 
>to do any of this work; the loader will do it all and pass it in to the 
>kernel in a format it can use directly.

Sounds excellent.  (Not that I'm offering at this point in time).

Initialisation code in the permanently resident kernel is basically
just wasting physical RAM.  The less of it there is, the better.
How much of the probe code can we move into userland?

Peter


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message