* Re: [9fans] why do we need 9load?
@ 2003-09-22 6:09 C H Forsyth
0 siblings, 0 replies; 6+ messages in thread
From: C H Forsyth @ 2003-09-22 6:09 UTC (permalink / raw)
To: 9fans
>>Seems to me I could load Plan 9 from flash, and then just have it load the
>>real kernel we want.
it was done on some older Plan 9 platforms.
it's also what we do with Inferno on some (not enough) platforms.
the kernel has got most of the drivers, after all.
if you keep a collection of commands in #/, you can do
flash management conveniently through a name space.
one of the dangers in spending too long on a bootstrap
is that you'll end up writing an operating system, with
its own drivers, its own command structure of course, with less
thought than you'd put into a proper one.
it's often as much code than the thing it's booting, or
even more, and certainly that much more to port and maintain.
you sometimes do need something small during the initial
stages of a port to a platform, especially if you must
load it as S records, but it's best not to go overboard.
9load is probably at the limit.
in other words, you realise 9load is not too bad,
once you've had to look at one of the sourceforge alternatives
and weighed it.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [9fans] why do we need 9load?
@ 2003-09-22 5:18 Ronald G. Minnich
2003-09-22 5:22 ` Russ Cox
0 siblings, 1 reply; 6+ messages in thread
From: Ronald G. Minnich @ 2003-09-22 5:18 UTC (permalink / raw)
To: 9fans
OK, I just a little while back RTFM (well, Andrey Read The FM, and said
"Ron, why don't you ever R The FM", so I did) and noticed /dev/reboot.
Hey, plan 9 can reboot Plan 9 kernels. Hey, the code is really slick. Why
can't I write code this good any more? Too old? Time for me to move into
management or something?
So now I'm wondering, after spending a few months jiggering 9load, why do
we need it? Could we use Plan 9 to boot Plan 9? It seems like it ought to
be possible, and would remove the burden of keeping 9load up to date
(which was a bit of work for the Geodes and CF).
Seems to me I could load Plan 9 from flash, and then just have it load the
real kernel we want.
Inquiring minds, minds which on rare occasions RTFM, want to know. What
I'd really like to do is test this out on EPIA, the currently really cool
cheapo mobo.
ron
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] why do we need 9load?
2003-09-22 5:18 Ronald G. Minnich
@ 2003-09-22 5:22 ` Russ Cox
2003-09-22 5:25 ` Ronald G. Minnich
0 siblings, 1 reply; 6+ messages in thread
From: Russ Cox @ 2003-09-22 5:22 UTC (permalink / raw)
To: 9fans
> OK, I just a little while back RTFM (well, Andrey Read The FM, and said
> "Ron, why don't you ever R The FM", so I did) and noticed /dev/reboot.
> Hey, plan 9 can reboot Plan 9 kernels. Hey, the code is really slick. Why
> can't I write code this good any more? Too old? Time for me to move into
> management or something?
>
> So now I'm wondering, after spending a few months jiggering 9load, why do
> we need it? Could we use Plan 9 to boot Plan 9? It seems like it ought to
> be possible, and would remove the burden of keeping 9load up to date
> (which was a bit of work for the Geodes and CF).
that was the plan. like many projects, it got stalled for lack of manpower.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] why do we need 9load?
2003-09-22 5:22 ` Russ Cox
@ 2003-09-22 5:25 ` Ronald G. Minnich
2003-09-22 17:52 ` jmk
0 siblings, 1 reply; 6+ messages in thread
From: Ronald G. Minnich @ 2003-09-22 5:25 UTC (permalink / raw)
To: 9fans
On Mon, 22 Sep 2003, Russ Cox wrote:
> that was the plan. like many projects, it got stalled for lack of manpower.
ok, so there's nothing insane about it. I'm going to give it a try and see
how it goes. It's easier to fix Plan 9 for linuxbios, then to fix both
9load and Plan 9 for linuxbios -- my calculator claims it is twice as easy
to fix half the number of things.
ron
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] why do we need 9load?
2003-09-22 5:25 ` Ronald G. Minnich
@ 2003-09-22 17:52 ` jmk
2003-09-23 15:14 ` ron minnich
0 siblings, 1 reply; 6+ messages in thread
From: jmk @ 2003-09-22 17:52 UTC (permalink / raw)
To: 9fans
On Mon Sep 22 01:26:56 EDT 2003, rminnich@lanl.gov wrote:
> On Mon, 22 Sep 2003, Russ Cox wrote:
>
> > that was the plan. like many projects, it got stalled for lack of manpower.
>
> ok, so there's nothing insane about it. I'm going to give it a try and see
> how it goes. It's easier to fix Plan 9 for linuxbios, then to fix both
> 9load and Plan 9 for linuxbios -- my calculator claims it is twice as easy
> to fix half the number of things.
>
> ron
I still tinker with it from time to time. There are some things about the
current /dev/reboot which I don't like.
It still doesn't solve the problem (for us great unwashed) of getting the
first kernel into memory from somewhere other than flash without writing
more 16-bit code without a real assembler.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] why do we need 9load?
2003-09-22 17:52 ` jmk
@ 2003-09-23 15:14 ` ron minnich
0 siblings, 0 replies; 6+ messages in thread
From: ron minnich @ 2003-09-23 15:14 UTC (permalink / raw)
To: 9fans
On Mon, 22 Sep 2003 jmk@plan9.bell-labs.com wrote:
> I still tinker with it from time to time. There are some things about the
> current /dev/reboot which I don't like.
one thing that would be real nice would be to seperate the loading of the
new image from the boot.
So I could:
echo /386/9pctest > /dev/rebootfile
disk/kfscmd sync
disk/kfscmd halt
echo reboot > /dev/reboot
Clean shutdown/restart. Anyway I'll be looking at this quick mod if nobody
beats me to it (or it's already done and once again I have failed to R The
FM).
ron
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2003-09-23 15:14 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-09-22 6:09 [9fans] why do we need 9load? C H Forsyth
-- strict thread matches above, loose matches on Subject: below --
2003-09-22 5:18 Ronald G. Minnich
2003-09-22 5:22 ` Russ Cox
2003-09-22 5:25 ` Ronald G. Minnich
2003-09-22 17:52 ` jmk
2003-09-23 15:14 ` ron minnich
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).