boot params

boot params

[Giuliano Pochini]

I'm trying to make the new aic7xxx driver work on my blue G3 and I
need to pass boot parameters to the kernel. I'm using yaboot. How
can I do ? Am I the only one who has problems with the new driver ?

Bye.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: boot params

[Ethan Benson]

[-- Attachment #1: Type: text/plain, Size: 556 bytes --]

On Thu, Apr 12, 2001 at 10:18:58PM +0200, Giuliano Pochini wrote:
> 
> I'm trying to make the new aic7xxx driver work on my blue G3 and I
> need to pass boot parameters to the kernel. I'm using yaboot. How
> can I do ? Am I the only one who has problems with the new driver ?

in your image= section:

	append="whatever"

and rerun ybin.  

which new driver are you referring to?  in 2.4?  i have never had to
do anything special to get this driver working on my blue g3 with a
2930.  

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

[-- Attachment #2: Type: application/pgp-signature, Size: 0 bytes --]

Re: boot params

[Giuliano Pochini]

>> I'm trying to make the new aic7xxx driver work on my blue G3 and I
>> need to pass boot parameters to the kernel. I'm using yaboot. How
>> can I do ? Am I the only one who has problems with the new driver ?
>
> in your image= section:
>
>       append="whatever"

Ok.

> and rerun ybin.

Do you rerun ybin ?  I just change yaboot.conf and it works.

> which new driver are you referring to?  in 2.4?

The new aic7xxx (v6.1.x).

> i have never had to
> do anything special to get this driver working on my blue g3 with a
> 2930.

I have problem with devices running at different clocks on the same bus.
v6.1.11 hangs during bus scanning.


Bye.
    Giuliano Pochini ->)|(<- Shiny Network {AS6665} ->)|(<-


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: boot params

[Ethan Benson]

[-- Attachment #1: Type: text/plain, Size: 506 bytes --]

On Fri, Apr 13, 2001 at 10:38:37AM +0200, Giuliano Pochini wrote:
> 
> Do you rerun ybin ?  I just change yaboot.conf and it works.

i do not screw with the bootstrap partition directly. its too much
trouble and will just get corrupted.  i edit /etc/yaboot.conf and run
ybin.

> I have problem with devices running at different clocks on the same bus.
> v6.1.11 hangs during bus scanning.

interesting.. never had any problems like that..

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

[-- Attachment #2: Type: application/pgp-signature, Size: 0 bytes --]

Re: boot params

[Ethan Benson]

[-- Attachment #1: Type: text/plain, Size: 653 bytes --]

On Mon, Apr 16, 2001 at 01:06:25AM -0400, Karim Yaghmour wrote:
> Ethan Benson wrote:
> > 
> > i do not screw with the bootstrap partition directly. its too much
> > trouble and will just get corrupted.  i edit /etc/yaboot.conf and run
> > ybin.
> 
> How does it get "corrupted"? I've been using hcopy to move things
> around and haven't had any problems. Are you suggesting that problems
> may arise from this type of manipulation?

the kernel HFS driver is unreliable.  hfsutils is fine, and thats what
ybin uses.  ybin is just far more convenient then doing everything
manually.  

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

[-- Attachment #2: Type: application/pgp-signature, Size: 0 bytes --]

Re: boot params

[Karim Yaghmour]

Ethan Benson wrote:
>
> i do not screw with the bootstrap partition directly. its too much
> trouble and will just get corrupted.  i edit /etc/yaboot.conf and run
> ybin.

How does it get "corrupted"? I've been using hcopy to move things
around and haven't had any problems. Are you suggesting that problems
may arise from this type of manipulation?

Cheers,

Karim

===================================================
                 Karim Yaghmour
               karym@opersys.com
      Embedded and Real-Time Linux Expert
===================================================

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: boot params

[Karim Yaghmour]

Ethan Benson wrote:
>
> the kernel HFS driver is unreliable.  hfsutils is fine, and thats what
> ybin uses.  ybin is just far more convenient then doing everything
> manually.

Oh, OK. Thanks for the info :)

Karim

===================================================
                 Karim Yaghmour
               karym@opersys.com
      Embedded and Real-Time Linux Expert
===================================================

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: boot params

[Iain Sandoe]

> Ethan Benson wrote:
>> the kernel HFS driver is unreliable.

Hmmm.  What do you mean?  I use it a lot (without problems AFAICT) - is
there something I should know?

ciao,
Iain.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: boot params

[Ethan Benson]

[-- Attachment #1: Type: text/plain, Size: 698 bytes --]

On Mon, Apr 16, 2001 at 12:38:17PM +0100, Iain Sandoe wrote:
> 
> > Ethan Benson wrote:
> >> the kernel HFS driver is unreliable. 
> 
> Hmmm.  What do you mean?  I use it a lot (without problems AFAICT) - is
> there something I should know?

apparently people get occasional filesystem corruption when copying
larger files around.  its always been rather shady since nobody really
gives a damn about it.  which is why something as important as the
kernels should not live on HFS.  and bootloaders should live on a
never mounted tiny HFS partition managed by hfsutils or preferably
ybin.  

it comes up on irc here and there.  

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

[-- Attachment #2: Type: application/pgp-signature, Size: 0 bytes --]

Re: boot params

[Iain Sandoe]

On Mon, Apr 16, 2001, Ethan Benson wrote:
> On Mon, Apr 16, 2001 at 12:38:17PM +0100, Iain Sandoe wrote:
>> > Ethan Benson wrote:
>> >> the kernel HFS driver is unreliable.
>>
>> Hmmm.  What do you mean?  I use it a lot (without problems AFAICT) - is
>> there something I should know?
>
> apparently people get occasional filesystem corruption when copying
> larger files around.  its always been rather shady since nobody really
> gives a damn about it.

Those of us with OW machines - booting with BootX (and a vestigial need to
use MacOS for 'rent-paying-work' ;-) _do_ give a damn.  I'd be interested if
someone could point at a reliable way to replicate a fault.

> which is why something as important as the
> kernels should not live on HFS.

OW machines don't have the choice.

>  and bootloaders should live on a
> never mounted tiny HFS partition managed by hfsutils or preferably
> ybin.

...

> it comes up on irc here and there.

Yes, but I've not seen a report for months - the last IIRC were for 2.3.xx
kernels ...

ciao,
Iain.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: boot params

[Ethan Benson]

[-- Attachment #1: Type: text/plain, Size: 782 bytes --]

On Mon, Apr 16, 2001 at 12:59:03PM +0100, Iain Sandoe wrote:
> 
> > which is why something as important as the
> > kernels should not live on HFS.
> 
> OW machines don't have the choice.

yes they do, its called quik.  

but for bootx it is still advisable to use hfsutils or only mount the
exchange partition rather then the real macos boot partition.
(exchange partitions used in a `/tmp' like capacity would not pose a
big deal to filesystem corruption)

> Yes, but I've not seen a report for months - the last IIRC were for 2.3.xx
> kernels ...

jeramy was mentioning some reports recently.  

in any event are you saying the hfs driver is even resembling being
close to bug free? 

didn't think so.

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

[-- Attachment #2: Type: application/pgp-signature, Size: 0 bytes --]

Re: boot params

[Giuliano Pochini]

> > > which is why something as important as the
> > > kernels should not live on HFS.
> >
> > OW machines don't have the choice.
>
> yes they do, its called quik.

Not a good choice. Often it happens you can't type anything in the
OF boot menu script, so you can't choose what OS start up (or at
least my 7300 with OF v1.0.5 had this problem). It also happens on
a 2nd-half-2000-series G4 where the ofboot script refuse to accept
keyboard input 80% of the times.

> > Yes, but I've not seen a report for months - the last IIRC were for 2.3.xx
> > kernels ...
>
> jeramy was mentioning some reports recently.

No problems here.

Bye.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: boot params

[Ethan Benson]

[-- Attachment #1: Type: text/plain, Size: 501 bytes --]

On Mon, Apr 16, 2001 at 08:11:21PM +0200, Giuliano Pochini wrote:
> 
> Not a good choice. Often it happens you can't type anything in the
> OF boot menu script, so you can't choose what OS start up (or at
> least my 7300 with OF v1.0.5 had this problem). It also happens on
> a 2nd-half-2000-series G4 where the ofboot script refuse to accept
> keyboard input 80% of the times.

everyone i know running quik doesn't have any problems.  

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

[-- Attachment #2: Type: application/pgp-signature, Size: 0 bytes --]

Re: boot params

[Iain Sandoe]

> On Mon, Apr 16, 2001 at 08:11:21PM +0200, Giuliano Pochini wrote:
>>
>> Not a good choice. Often it happens you can't type anything in the
>> OF boot menu script, so you can't choose what OS start up (or at
>> least my 7300 with OF v1.0.5 had this problem). It also happens on
>> a 2nd-half-2000-series G4 where the ofboot script refuse to accept
>> keyboard input 80% of the times.
>
> everyone i know running quik doesn't have any problems.

That may be true with quik itself - but console usage of OF is a PITA for
most of the older (OF 1.0.5-ish) machines - I can't get my 9600 to retain
the re-mapping of kbd/screen for more than one reboot - i.e. set the OF vars
(using BootVars) reboot - OF is OK to screen & kbd.   Next reboot they are
reset back to point to 38400 serial :-((( ...

BootX is much nicer... :-)  especially for those of us who still _need_ to
use MacOS for other work...

ciao,
Iian.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: boot params

[Michel Lanners]

On  17 Apr, this message from Iain Sandoe echoed through cyberspace:
>
>> everyone i know running quik doesn't have any problems.
>
> That may be true with quik itself - but console usage of OF is a PITA for
> most of the older (OF 1.0.5-ish) machines - I can't get my 9600 to retain
> the re-mapping of kbd/screen for more than one reboot - i.e. set the OF vars
> (using BootVars) reboot - OF is OK to screen & kbd.   Next reboot they are
> reset back to point to 38400 serial :-((( ...

You sure your backup battery is OK? It's not completely dead, I think
otherwise the box wouldn't boot; but you may test it anyhow.

Also, you could add an nvsetenv script on shutdown as a workaround.

Michel

-------------------------------------------------------------------------
Michel Lanners                 |  " Read Philosophy.  Study Art.
23, Rue Paul Henkes            |    Ask Questions.  Make Mistakes.
L-1710 Luxembourg              |
email   mlan@cpu.lu            |
http://www.cpu.lu/~mlan        |                     Learn Always. "


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: boot params

[Ethan Benson]

[-- Attachment #1: Type: text/plain, Size: 537 bytes --]

On Tue, Apr 17, 2001 at 04:03:45PM +0200, Michel Lanners wrote:
> 
> You sure your backup battery is OK? It's not completely dead, I think
> otherwise the box wouldn't boot; but you may test it anyhow.
> 
> Also, you could add an nvsetenv script on shutdown as a workaround.

well on some machines the MacOSROM resets the nvram, so you must never
allow it to boot.  quik is generally best for people wanting to get
rid of the useless space wasting MacOS in these cases.  

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

[-- Attachment #2: Type: application/pgp-signature, Size: 0 bytes --]

Re: boot params

[Iain Sandoe]

On Tue, Apr 17, 2001, Ethan Benson wrote:
> On Tue, Apr 17, 2001 at 04:03:45PM +0200, Michel Lanners wrote:
>>
>> You sure your backup battery is OK? It's not completely dead, I think
>> otherwise the box wouldn't boot; but you may test it anyhow.
>>
>> Also, you could add an nvsetenv script on shutdown as a workaround.
>
> well on some machines the MacOSROM resets the nvram, so you must never
> allow it to boot.

I think this is the case with 9600 - it's always behaved this way and I've
no (other) reason to be suspicious of the battery (you usually get a warning
in that several power button presses are required to start it).

Anyway all my machines power are on & active 24/7 - even those running MacOS
;-?

One day I might try to figure out something that will work - because it _is_
possible to change the vars using BootVars and get one 'reboot' into OF
screen/kbd with the new params.

Iain

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: boot params

[Timothy A. Seufert]

At 1:54 PM -0800 4/16/01, Ethan Benson wrote:
>On Mon, Apr 16, 2001 at 08:11:21PM +0200, Giuliano Pochini wrote:
>>
>>  Not a good choice. Often it happens you can't type anything in the
>>  OF boot menu script, so you can't choose what OS start up (or at
>>  least my 7300 with OF v1.0.5 had this problem). It also happens on
>>  a 2nd-half-2000-series G4 where the ofboot script refuse to accept
>>  keyboard input 80% of the times.
>
>everyone i know running quik doesn't have any problems.

It's not a good choice for all machines though.  It can be *very*
difficult to get working in some of Apple's more broken OF
implementations.  (beige G3 for example)

Or at least that's how it was last I tried.  I've not tried to use
quik for over a year, for just that reason.

   Tim Seufert

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: boot params

[Ethan Benson]

[-- Attachment #1: Type: text/plain, Size: 707 bytes --]

On Tue, Apr 17, 2001 at 11:15:16PM -0700, Timothy A. Seufert wrote:

> It's not a good choice for all machines though.  It can be *very*
> difficult to get working in some of Apple's more broken OF
> implementations.  (beige G3 for example)

funny, http://penguinppc.org is booting from quik.  there is some
flakyness it seems you need to put some random crap in boot-file to
keep quik from crashing (its not used as an argument though).  

> Or at least that's how it was last I tried.  I've not tried to use
> quik for over a year, for just that reason.

well it helps to use a non-broken version of quik, such as the one in
debian.  

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

[-- Attachment #2: Type: application/pgp-signature, Size: 0 bytes --]